Go to the Globe and Mail homepage

Jump to main navigationJump to main content

B.C. NDP Leader Adrian Dix walks past an exit sign during a tour of Stoneboat Vineyards during a provincial election campaign stop in Oliver, B.C., on May 3, 2013. (Darryl Dyck/The Canadian Press)
B.C. NDP Leader Adrian Dix walks past an exit sign during a tour of Stoneboat Vineyards during a provincial election campaign stop in Oliver, B.C., on May 3, 2013. (Darryl Dyck/The Canadian Press)

Voters who remained undecided until election day a main factor in B.C. result: pollster Add to ...

While the reliability of pollsters has been under the microscope in the days following the B.C. election, one polling firm says new number tells part of the story of why everyone was predicting an NDP victory in the final days leading up to the election, and why the Liberals were able to pull off a victory.

More Related to this Story

According to an election-day poll conducted by Ipsos Reid and released the day after the election, one in 10 B.C. voters decided who they would vote for on election day, and many of these voters cast ballots for the Liberals.

The firm also said that nearly one in five voters said they decided who they would vote for in the last week of the campaign. Forty-one per cent of these voters ended up voting Liberal, while 34 per cent voted for the NDP.

The firm said that when you couple this with a low voter turnout – 52 per cent – and the fact that many NDP voters did not fulfill their promise to vote and instead stayed home, the Liberals were able to defy pollsters and pundits.

But Ipsos Reid said voters who decided last minute can only partially explain the discrepancy between pre-election polls and the final result, and the verdict is still out on what went wrong.

“This is certainly not the case of everyone making up their minds at the last minute and that’s why we’re wrong,” said Kyle Braid, a spokesperson for Ipsos Reid. “There still some factors that us, as an industry, need to identify. At some point, we’re going to have check with the other players and see if we can come to some collective agreement as to what happened.”

Mr. Braid said what is clear already is that pollsters need to rework their models to better determine if the people they’re surveying will actually turn out to vote.

“If the profile of who turns out to vote is different than the overall population, then the results can be off,” he said.

Follow on Twitter: @danbitonti

In the know

Most popular video »

Highlights

More from The Globe and Mail

Most Popular Stories