Here in Manitoba, it takes a little longer for news from Alberta to make its way up the swollen Red River to Winnipeg, where the major story of the day is whether the town’s new NHL hockey team is to be named the delightfully retro Jets, or something new.
So it was rather late when we learned the startling news that back in Alberta the Wildrose Alliance is within “striking distance” of the governing Conservatives, and according to a public opinion poll, no less.
Well, maybe… But the poll on which these reports are based needs to be considered with caution, and quite possibly taken with the traditional grain of salt. There are many questions that must be answered about the poll before it can be taken seriously.
The Calgary Herald reported on June 14 that the online poll commissioned by the Wildrose Alliance and done by Abingdon Research indicated 29 per cent of decided voters “are either supporting or leaning towards Danielle Smith’s Wildrose Alliance.” In a blog post, an Edmonton Journal writer noted that the poll had been released by party executive director Vitor Marciano, who made the claim, “we are closing in on them.” (The Tories, that is.) “We are within striking distance.”
While 34 per cent of the poll’s respondents either support or lean toward the Progressive Conservatives, as the Herald reported, it does show the Alliance with more support than some earlier polls. The poll was conducted between May 20 and May 24, 2011, and 2,433 respondents’ answers were considered.
However, several questions immediately arise:
Why is there absolutely no evidence cited in the Herald’s story to back up the claim in its headline that “New Wildrose commissioned poll shows Wildrose gaining on Tories.” Just an observation, but back in the day it was considered standard operating procedure in journalism to support the claims in your headlines and leads with verifiable facts. Marciano’s opinion, which was not even quoted in the Herald, should not be enough.
Can one be a “decided voter” and only be “leaning toward” supporting a particular political party? Maybe this is just me — after all, I used to be an editor — but what’s with this?
It’s an online poll. So how were the poll’s respondents selected? No one — other than the pollster, presumably — has any idea how biased or unbiased this panel is, although it is fair to say that on-line polling is not as trustworthy a method as randomly selected telephone calls.
So is there a systemic bias in the sample? One hint may be in the poll’s assessment of how many committed voters polled have an opinion about Conservative leadership candidate Rick Orman — 35 per cent. Could 35 per cent of a truly random sample of Alberta voters even know who Rick Orman is? It seems unlikely to this blogger — unless this is a sample biased toward conservative and ultra-conservative voters who would be aware of Orman’s as-close-to-the-Wildrose-as-possible platform.
Finally, who or what is Abingdon Research? Well, here’s what we know: the company has no known association with any reputable polling or marketing association. It specializes in political strategy and does such things as identify support for conservative parties.
Abingdon’s principal research officer is a fellow named Hamish Marshall, who is identified on a website called “Conservative Home-Centre Right” as “a well known strategist and activist trainer within Conservative circles” and “an active Conservative in Canada.”
The Marshalls, along with Marciano, also all seem to have an interesting association with Calgary West MP Rob Anders, who was born on April Fools Day 1972 here in the ‘Peg, was the sometime director of director of the Astroturf organization Canadians Against Forced Unionism and is best known for calling Nelson Mandela a “terrorist.”
Is a picture beginning to form here? While it is possible that the Abingdon poll is legitimate, without more information it would also be reasonable to conclude that this might be a poll that was administered to a self-selecting sample for a client that desired a specific outcome.
It is interesting that the release of the Wildrose Alliance claims followed by only four days the publication in journalist Paul McLaughlin’s subscription-only Alberta Scan newsletter the details of a scientifically valid poll with results much less favourable to the Wildrose Alliance.
That telephone survey of 900 Albertans conducted between March 14 and April 3 by Trend Research and commissioned by McLoughlin and pollster Janet Brown, “showed 50 per cent of those intending to vote would have voted for the PCs, up from 40 per cent in January. The Wildrose would have received 22 per cent, down from 33 per cent…”
In other words, according to that poll, once Premier Ed Stelmach said he was on the way out, Tory support jumped 10 per cent and Wildrose support plummeted by the same amount.
In fairness, the Trend poll was conducted more than two months before the Abingdon effort, so they may have been measuring a different mood. But there are nevertheless strong reasons not to bet the farm on the results of the Abingdon poll.
It is genuinely shocking, moreover, that the mainstream media has reported its results so uncritically despite its obvious whiff of Wildrose propaganda.
If the Calgary Herald or the Edmonton Journal wish to know what Albertans really think now about the Wildrose Alliance and the other Alberta political parties, they ought to commission a scientific poll of their own.
This post also appears on David Climenhaga’s blog, Alberta Diary. Thank you for choosing rabble.ca as an independent media source. Rabble is a reader-supported site — visited by over 315,000 unique visitors during the election campaign! But rabble.ca needs money to grow. Support rabble.ca as a paying member or by making a one-off donation .