So far today I've seen at least two different bloggers base their posts on this poll-based article (though Wells links to it in a different CanWest paper) as if we can draw definitive conclusions from it. If you check you'll find the story doesn't reproduce the exact questions asked in the poll nor are there internals supplied. And if you drop by the Ipsos Reid website you'll be politely informed that the information in question is "Premium Content" and you must be a subscriber to see it.
It gets silly enough to watch the news media themselves commission these polls — creating their own news — and then report on what they supposedly tell us about public opinion without acknowledging said media's own role in creating that opinion. But it's even sillier when we can't see exactly what the poll results really mean. Believe it or not even a poll that originates as an honest attempt to measure opinion rather than a push poll designed to manipulate it can still ask questions in such a way as to render it less than useful. And besides, measuring a new Liberal leader's popularity without telling us the regional distribution of results in a country like Canada really doesn't tell us much.
I find it difficult to get excited about stories like these. So usually I don't.
PS:
On the other hand, allow me to join a number of bloggers who are encouraging everyone to read this Paul Wells piece on Stephen Harper.
I agree on one point, if you're releasing a PUBLIC poll, then you should make the complete results PUBLIC.
I'd go further than that. If you're a media company basing stories on polls you yourself have commissioned, presenting those stories as news and then expecting people to subscribe to the polling company's site to see what the real results are, then people should be pointing at you and laughing. Instead you're treating them like they're, you know, serious about something other than manipulating opinion and making more money for the polling company. If we accept this nonsense, they'll keep doing it. You get the behaviour you reward. You discourage the behaviour you punish.
Having said all that, of course I've done it too but I'm trying to train myself to stop.
You know the last few times I've seen a hyped poll in Canadian politics I've immediately thought, well, crap, no Nate Silver, so I'll go look at the questions and sponsor myself, only to, wham, walk into the glass doors of the polling co websites.
This stuff's embarrassing. If the poll's questions, methods, sponsor and sampling date aren't available, it isn't worth jack all. Can the high school AV club we call a media maybe take note?
One earlier ipsos reid poll had questions that I took issue with. For example:
"Unholy deal". Really?
Source: (in the "Detailed Tables" link) http://www.ipsos-na.com/news/pressrelease.cfm?id=4201
Paul Well's article on Harper is a must read for all. Make no mistake, Harper does not want to work with anybody and hates govt.
I think Well's would support a coalition govt ~ just to make all this craziness end and get some stable govt.
I forgot, by the way, my favourite oddity of recent polling - the CBC/EKOS poll, the first 'bad' one on the coalition, which asked Canadians if they thought parliamentarians should "take a break."
This should have been Charles I's approach. "No, no, we were just taking a break!"
I'd missed that one entirely. Does this mean recess will become a regular feature of parliament — 15 minutes in the morning and 15 minutes in the afternoon? Will they have teachers on duty in the schoolyard to make sure everyone behaves?
Pogge, school has changed, and in Ontario we have nutritional breaks, 3 of them, that are equal in length. We do not have recess and in elementary school there is not enough time to go home for lunch, so all the kids stay (unless they live within 5 minute walking distance from school),
So I guess I showed my age, huh?