Politicians are fond of saying that 'the only poll that matters is the one on election day.'
Gosh darn it — they might be right.
After embarrassing flops in both the Alberta and Quebec elections, it looks like Canada's pollsters got 'it' wrong, again, with the B.C. election.
The final opinion polls, released Monday, suggested that the NDP would win a majority government with a six point lead on the Liberals in terms of the popular vote.
That didn't happen: As it turns out, Christy Clark's Liberals won a majority with a five point lead on the NDP.
[ Related: Liberals pull off surprise victory in B.C. election ]
Pundits, analysts and bloggers certainly need to assume some of the responsibility for publishing polls without proper explanation. But, on Tuesday night/Wednesday morning, the public scorn was reserved for the pollsters.
There are certainly several challenges to the polling industry: Cell phones, caller ID's, and lower voter turnout levels, to name a few. But how could they have gotten this so wrong? Didn't America's Nate Silver successfully predict all 50 states in the presidential election last Fall?
A couple of pollsters offered their theories on Tuesday evening.
Polling analyst Eric Grenier gave this explanation on his website:
The results will fall within the high and low forecasted outcomes, but they were considered to be an unlikely event. It seems, instead, that with the polling we have in Canada we can expect these sorts of surprises more often. It is very disappointing.
Ipsos Reid's Darrell Bricker also 'faced the music' on Twitter:
In a recent op-ed column for Postmedia News, Peter Loewen, an assistant professor of political science at the University of Toronto, suggested that pollsters cannot be certain of the accuracy of new technologies such as online and text message polling.
Pollsters simply do not know enough about who responds to polls via some media, who replies through others, and what kinds of people ignore polling requests entirely.
The problem isn't getting sample, it's getting good sample. Simply knowing a respondent's demographic information is not enough to correct for bad sampling. The result is that we cannot extrapolate with sufficient accuracy from our samples to the whole population.
We cannot, in other words, know with much confidence the likely outcome of an election before the votes are cast.
Kudos to Grenier and Bricker for not going into hiding.
But the B.C. election is clearly another black-eye for the industry.
Are you a politics junkie?
Follow @politicalpoints on Twitter!