Polls, damned polls and questions

A lot depends on what you ask:

As the Canadian death toll climbs in Afghanistan, conventional wisdom would suggest that public support for the bloody mission will plummet in direct response.

Polling data accumulated over the past year and a half, however, tells a more complex tale, indicating that opinion on the divisive issue has held relatively stable -- sometimes even after troop deaths -- and that Canadians may be more likely to approve of the historic military mission when they are told more about it.

"It's been incredibly consistent," said John Wright of pollster Ipsos Reid. "We've polled during some of the worst times for the Canadian military, we've been in the field when there have been six soldiers killed ... We've been sure we can [conduct polls] whenever sentiment would be worst, and it seems to have held."

In fact, what pollsters ask people would appear to have almost as much impact on opinion as what is happening in Afghanistan itself [emphasis added], some analysts say.

When questions in a Defence Department poll emphasized protecting civilians and rebuilding the country, support for the mission shot up.

When asked by Decima Research if they thought the number of Canadian casualties was acceptable, on the other hand, two-thirds of respondents answered in the negative.

The deaths of soldiers from the Quebec-based Van Doos regiment could alter the whole equation. In the one province already firmly opposed to sending troops to Afghanistan, a CROP survey partly conducted after the death of Private Simon Longtin on Sunday recorded an 11 percentage point increase, to 68%, of Quebecers opposed to their compatriots being involved in the conflict. That was before the two most recent deaths.

But until the past few days, at least, opinion levels were surprisingly predictable.

A series of polls conducted by Ipsos Reid for CanWest News Service and Global Television since January, 2006, all asking the same question, has seen support for the mission roller-coaster from 44% to 52% [an eight percent variation is a "roller'coaster"?], then back down below 50%, then up again to a peak of 57% last fall. There has been a slow slide to 50% support since then, but the results over 18 months plot a relatively flat line that has hovered around 50% backing...

Strategic Counsel has recorded similar fluctuations and a similar range of variation in its polls since early 2006, after a fast drop from 55% support in March of that year.

The difference is that its surveys have backing for the mission hovering around the 40% mark, 10 points below those of Ipsos Reid.

A spokesman for the company refused to comment on its results, citing its contract with another media outlet. Rob Huebert of the University of Calgary's Centre for Military and Strategic Studies suggested the difference in the two pollsters' results can be traced to the questions they pose.

Ipsos Reid asks respondents about their support for "the use of Canada's troops for security and combat efforts against the Taliban and al-Qaeda in Afghanistan."

Strategic Counsel asks simply about "the decision to send Canadian troops to Afghanistan."

"They set up the issue differently," Prof. Huebert said. "It is not a coincidence that the one that clearly defines the threat gets the higher response ... People won't think immediately of why we are there. If you mention Taliban and al-Qaeda, people will clue in."

In a string of polls for the National Defence Department in late 2006 and early 2007, Ipsos Reid tweaked the questions even further, and found dramatic differences in response.

When the question referred to military operations that help to secure "the environment for the civilian population" through activities "that include combat," backing shot up to the low 60s.

After a lengthy preamble that said Canada is trying to improve human rights for women and build a more free and democratic society in Afghanistan, another question drew support from 81%.

It suggests that explaining the purpose of the mission is all-important if the government wants to boost the tepid support among Canadians for the operation, said Alex Morrison of the Canadian Institute for Strategic Studies.

"The polling has indicated to the government that it is not doing a good job of communicating why we are in Afghanistan. The government recognizes that but consistently refuses to do anything about it," he said.

"Unless the government starts telling Canadians often, frequently, why we are there, I don't think the numbers will change very much."

Latest:

Canadian support for the mission holding steady: poll
Albertans at 72% are most likely to back deployment, while nearly two out of three Quebecers are opposed