You’re going to see a lot of polling data thrown around during this election, and wonder what it all means. As we’ll see, that’s kind of by design—just another one of the many “behind closed doors” aspects of Canadian politics.
To illustrate the conundrum, I’ll tell one of my favourite Jewish jokes, one that I’ve told on campaign many times, and which absolutely kills in every region of Ontario—even (one time) when translated into Mandarin:
A Jew is stranded on a desert island. When the rescue team comes to save him, they see that he has not only survived, but prospered: he grows his own crops, raises his own animals… he’s even built two synagogues on the island. Two synagogues? Why build two, if he’s the only person on the island? Well, he says, I pray in that one, and I wouldn’t set foot in the other one if you paid me!
And so it goes with polling. The numbers are there, but once they’re out, it’s anybody’s guess what the parties and the voters are going to do with them.
For example, a provincial election concluded in Nova Scotia last. If you read the headlines, the pollsters missed the mark in a big way—a Liberal win was predicted, but the Progressive Conservative Association pulled out a decisive victory.
The following are all real things people have either said or written to explain away this error:
- “Well, they did capture the momentum behind the Nova Scotia PCs in the last few days…”
- “Not all the pollsters got it wrong. Pollster X was preeeeeeetty close!”
- “Polling in Canada is fine, if you recognize what it’s good for and what it’s not good for!”
OK. In that case, what was the result good for? Can we use the Nova Scotia election to predict how the big dance will turn out?
Probably not. There are under a million Nova Scotia residents, and 55 percent of those eligible cast a ballot. That works out to about twice the country’s Jewish population. Victory was decided by a few hundred votes in most ridings, which would be recount territory in central Canada.
And so, these poor numbers were consigned to the same fate as all polls done in this country—grist for whatever narrative interested parties wanted to push.
Outgoing premier Ian Rankin called an early election without explaining why he did so to voters, so Justin Trudeau, who did the same thing, will doubtlessly share Rankin’s fate.
Also, incoming premier Tim Houston says he isn’t even a member of the federal Conservative Party of Canada, and the N.S. PCs are much closer to the federal Liberals than the CPC, so this result is an endorsement of Trudeau, not a repudiation of him. Obviously.
Now, if the partisans spinning the numbers can be compared to the Jew who builds two synagogues only to reject one as unsuitable, the pollsters themselves are the synagogues.
This is because your basic Canadian polling outfit…
- has a charismatic and somewhat controversial founder and spokesperson
- usually has an under-the-radar feud simmering with the other pollsters
- is competing for the attentions of the same relatively small number of people
- only gets called into the spotlight during a particular time of the year
- will put aside any feud to band together against some new upstart trying to barge onto their turf
So, would I endorse any numbers coming out this early, in the summer—(when, the partisans tell us, nobody is paying attention?) Not if you paid me!
Josh Lieblein can be reached at [email protected] for your response to Doorstep Postings.
- Doorstep Postings #1: A different kind of Canadian federal election view from Josh Lieblein
- Doorstep Postings #2: An election lesson from John Tory’s faith-based schools fumble in Ontario
- Doorstep Postings #3: ‘Cringe’ is a word that explains many Canadian political campaigns