In Saturday's Globe and Mail, editor Edward Greenspon said his paper will investigate a poll that gave the ruling B.C. Liberals much more popular support than they eventually won in the election.

 The poll -- made public on May 13, four days before the election -- was made an issue by The Tyee in a May 16 article.

The poll, conducted by The Strategic Counsel (addendum: And co-sponsored by CTV), gave the Liberals a 13-point lead over the NDP.

In comparison, an Ipsos-Reid poll gave the Liberals an eight-point lead while the Mustel Group gave the Liberals a five-point lead.

The Liberals won the election, taking 46 per cent of the popular vote to 41 per cent for the NDP.

Here is an excerpt from the Tyee story:

On the Friday before the tightest provincial election facing British Columbia this decade, Canada’s most influential newspaper published dramatic poll numbers that appear to be based on biased methodology.

“This is unethical,” Vancouver pollster Angus McAllister told The Tyee. “The Globe and Mail ought to be responsible enough not to publish a poll where the ballot question is preceded by 14 questions that influence people’s answers.”

Here is what Greenspon wrote in response:

From my perspective, the questions asked were self-evidently neutral.
 
... Our poll did end up being outside the margin of error, and other polls proved to be more accurate, matters of concern both to The Globe and The Strategic Counsel.

In the end, the election vote tally for the Liberals did turn out to be within the poll's margin of error.

Since the NDP moved up five points and the Greens went down four, Greenspon wondered if  strategic voting could explain what happened -- or if it was simply a rogue poll:
Our survey may have been such a rogue, but it seem more likely that, as with the last federal election, voters made switches at the last moment, most probably to prevent another landslide by Premier Gordon Campbell's Liberals. With the diminution in recent years of traditional party loyalties, switching , and last-minute switching, have become far more common. That in itself is an issue worth our pondering.

The Strategic Counsel is currently conducting a re-examination of its data to try to figure out if we were victims of volatility or a rogue. We take the matter seriously.


Angus McAllister reached this conclusion:

“It’s the Globe’s responsibility,” McAllister concluded. “The Globe and Mail appears to have misrepresented a message-testing poll as a public opinion poll. And they’ve released those results just four days before an election. This is definitely unethical.”

At Antonia Zerbisias's blog, she noted the Globe answered some questions put to it by Tyee editor David Beers in a May 17 article.

Here is an excerpt from Globe managing editor Colin McKenzie:

Angus MacAllister's huffing and puffing in the piece would actually have some methodological merit if it did not apply to the questionnaire we administered. The bias associated with "structuring" and "framing" effects occur when preceding questions are argumentative or biased in one direction, only. Our questions, in every instance, give respondents an equal opportunity to answer in the affirmative or negative and therefore to support either the Liberal or NDP position.

By posing the questions we are trying to more closely approximate the dynamic that will occur in voters' minds in the final days before election day.

There was no 'message testing' involved, or it goes without saying,. any intent to do anything but report what British Columbians are thinking in the week before an election. All of Strategic Counsel, The Globe and Mail and CTV have a long history and great credibility about polling. It is outrageous to suggest we may be favouring one side or another in our news pages. It is so outrageous that one wonders about the motives of those make the charges.