Frances Bula header image 2
← →

B.C. pollsters find new tools to read our minds

May 3rd, 2010 · 6 Comments

When you cover politics in this province, you inevitably end up covering polling.

(Current interesting issue is Mayor Gregor Robertson’s numbers. I’ve been told consistently for the past year that his approval rating is at 70 per cent, which is very respectable in the politician world, but don’t know if that’s taken a hit in recent months.)

But the little-known fact about polling is how much its world is changing, just as the worlds of many other forms of communications with the public are changing. Like the mass media and music industries, polling companies also have to deal with the tendency for everyone to want free information. (Hence, many of the polls you see reported are polls they’ve done for free, mainly as marketing devices.)

They’re also making strides in the worlds of psychology (is this person really telling me the truth? how can I find out?) and social media to try to delve into what we really think. It’s a tricky world. As anyone who’s ever had to deal with a marriage, kids, employees, employers, or people on the bus will tell you, what people say they want/believe/have done or often at odds with what they really want/believe/have done.

I got to explore this world in a little more depth in BCBusiness this month, where pollster Angus Reid is exploring the online frontier, Evi Mustel is refining the ways to get people to give honest responses, and Greg Lyle is looking at the role that emotion and long-term values play in in how we think about the world (and what we say to pollsters as a result).

In spite of all that, as Bob Penner of Strategic Communications notes, they’re still not political analysts — even though they like to pretend they are.

Categories: Uncategorized

  • booge

    They canvas any of those having to live next door to chicken coops, bee-hives with killer bees, or front-yard veggie gardens covered in stinking manure?

    No… Didn’t think do.

  • Michael Geller

    Whenever I think about polling, or voting of any kind for that matter, I am often reminded of an experience I had at a Metro Vancouver workshop a couple of years ago. The purpose of event was to get opinions from a wide variety of people on various issues related to the region. Each participant was given one of those electronic devices that allows you to select choice A, B, C, etc. The results were then tabulated and posted on large screens so the participants could all see the results.

    After the first hour I began to notice a pattern. Choices A, B, and C were picked far more often than choices E, F or G. It got to the point where I was impressing the people at our table with my ability to predict the results, regardless of the issue. Once I pointed out why I thought a certain opinion would be most popular, they too realized it wasn’t very difficult.

    While I have not seen any statistical evidence, I suspect the same thing may happen in some elections when voters are faced with a lot of choices. Those candidates whose names begin in the top half of the alphabet are likely to get more votes than those whose names begin with ‘n’ and onwards. (That’s why Marty Zlotnick’s victories were always so impressive!)

    I would be interested in whether the professional pollsters agree with my suppositions. I would also welcome their commentary on a proposed solution…to ‘scramble’ the questionaires or ballots so that each question or candidate occupies the top (and bottom) position an equal number of times.

    With the aid of computers it would be very easy to print out different questionaires and ballots. And should I ever decide to run again, I won’t have to change my name to my mother’s maiden name…..Abbott!

  • A. G. Tsakumis

    Funny this…and to think I almost missed it.

    You were one of the first people to laugh at staff claims in 2006 that Sam was high in the polls as well–although it was at a time right after the Torino flag spinnerama, when his numbers may in fact have been quite good (even though his bizarre initiatives had started to seep out from under the mayor’s door).

    And yet, when you are fed horse manure from Gregor’s people, after they have been pummeled for their radical agenda, you just accept it and wonder (lament?) if his recent numbers have taken a hit…

    He is actually polling in the high 50s low 60s with high disapproval numbers–that’s why they will not release their numbers…

    But when it was Sam, you were poking around extensively, including asking me if I had seen the numbers and had a copy.

    Recently you actually had the temerity to write in another post that you somehow loathed (my word) those of us who speak approvingly of politicians, only to one day become disenchanted–thus critical,and that this is “juvenile” to you–as if this was some sort of predetermined state.

    Well, let me tell you what is really “juvenile” and actually quite troubling: to watch a woman who used to be the standard against which everyone else was measured, devolve into a sycophantic, breathless cheerleader, whose opinions of a runaway train council have become the very definition of agenda journalism.

    Oh, excuse me, I forgot, REAL journalism is also about putting out a post on its own showcasing a man who admits in the story that his wife may in fact be at fault in an auto accident with the Mayor of Surrey, but you featured his unsubstantiated and (RCMP) discredited statements–did I mention in a single post, all by its lonesome?

    Must be nice up there on that cloud…

  • Tessa

    A.G. Tsakumis – if you have a poll, cite it, give a link, provide details as to who did the poll, what the questions were, who it was for, etc. Without that information, you’re just typing numbers in a comment pad on a blog, and it’s not exactly reliable. Too often people accept numbers on a screen, not just on blogs but in the mainstream media as well, where too often reporters don’t ask all the questions. If you want someone who knows how to analyze polls, check Nate Silver’s blog at http://www.fivethirtyeight.com, though admittedly it’s focused on the States.

    As for polls, I would be one of those people not responding to most. I don’t mind doing opinion surveys for the City or for public organizations that serve me, but how am I bettering the world by telling some pollster how I’m going to vote? How do these polls improve our democratic process? I would argue they have the opposite effect, doing exactly as critics in the story said, in turning pollsters into semi-psychic political commentators who somehow know why everyone is responding the way they do when they never bothered to asked that question.

    As for market research, again I’d like to know why we willingly agree to let companies learn how to manipulate us into buying products we don’t actually need? What’s in it for us, really? I don’t believe that market research in any way improves the products, or makes it more relevant to our lives – all it does is attach emotions to those products that help them sell things when we wouldn’t buy them if we actually thought logically about it.

    Even polls done for governments I find can be problematic. Governing by poll results is a terrible method of governance, and tends to produce scattered governments lacking in any strong priorities.

    Let’s be honest with ourselves: what good has this multi-billion dollar industry done for society as a whole? You can argue it’s done wonders for elites – both in political and business realms – but beyond that, I don’t see much good.

  • A. G. Tsakumis

    I note for the record that you did not ask Frances to show us her evidence, just me…

    Interesting…

    LOL!

    I have an email from a Vision insider who has seen the numbers, minus the BS shoveled by Vision hierarchs…

    And this is a most credible source.

    If the Mayor’s numbers were good and it was a poll done by a credible, non-Vision related firm, then they would have been splattered across the front of the Vancouver Sun…or featured exclusively in either Frances’ or Allen Garr’s columns…

  • Frances Bula

    For the three people following this thread — the Visions, yes, were letting it be known to various people at the end of the first year that the mayor’s polling numbers were good. I didn’t buy the 80 per cent number that one person claimed, but the 70-per-cent number that I heard from someone I consider to be a credible source seemed reasonable to me. Not surprising for a new mayor with a new agenda who didn’t have any major blunders in his first year. (Chickens and community gardens might stir up a shit storm in certain circles, but didn’t register with the larger population in their earlier incarnations.) I would guess that Dianne Watts is around the same, though likely she’s taken a little dip too recently because of some of the controversy over the Vaisakhi parade.

    Vision didn’t seem to want to put the numbers out, possibly realizing that the inevitable second chapter would be a decline in the numbers — not surprising for any politician after the glow has worn off. (BTW, not too complicated what the question is: Do you approve of the mayor’s performance so far or something along those lines.)

    I haven’t been interested enough to pursue this too aggressively. It’s interesting but early days yet. The problem for all of us is that pollsters are not interested in doing polling on civic issues or personalities. While many of them do free polls on provincial and federal issues, the civic issues are too messy and fragmented. So we’re always left with various rumours from the local parties about approval ratings. Sometimes that actually even give us the poll, but recent past examples were so problematic with the way they worded things (e.g. the Sullivan and Ladner polls as they were battling for the nomination) that it was hard to make any sense of them.

    The reality is that we can never really know what the numbers are unless someone chooses to sit in the phone room the entire time and listen in on every phone call. Everyone has an agenda and anyone can just tell all of us interested questioners anything they like, really, or choose to report the results of one question and not another. As do all reporters who’ve been slogging in the trenches for a long time, you have to assess how trustworthy you think your source is and how reliable they’ve been over the years. There are people I talk to who I know have, shall we say, been less than forthcoming with the truth and I assess their information accordingly.