We ran three polls on electoral reform.
Three different wordings of the same question about proportional representation. Here's what we found.
On the night of Britain voting to leave the European Union, The Strategists had an emergency podcast to talk about the consequences of that decision and referenda in general.
Contrary to the popular belief that they are the purest form of democracy, referenda are the most gameable type of democracy. Complex issues are reduced to binary choices. Compromise is impossible. And most problematic of all: somebody has to pick the question. And by picking the question, you pick the outcome.
On the last point, we wanted to move past rhetoric and Monday morning quarterbacking. We decided to test just how much that might be the case here at home on a very real issue: switching from first-past-the-post elections to proportional representation. From July 5 to July 10 we ran three polls in tandem with three different wordings of a referendum question. How much would the wording of the question affect the outcome of the poll?
As revealed in Episode 582 (44:10 mark), it turns out a great deal. Our data is below.
Corey Hogan
WHAT WE DID We used Google Consumer Surveys to poll 1500 respondents on one of three different phrasings of a possible referendum question, resulting in 500 responses to each of the following: First, a caveat on what we weren't testing: the popularity of switching to proportional representation - if we were, this survey would fail on a number of standards.
Google Consumer Surveys are designed to get the opinions of the internet population and are weighted accordingly - this is not a model of eligible or likely voters. This survey was in English only. And, of course, proportionality is a concept that can be implemented in a number of ways: MMP, STV, list PR, etc... an actual referendum would likely ask about a specific model. We didn't.
What we were testing was how changes in language impact the results, and we controlled for time in field and audience composition.
"The right word may be effective, but no word was ever as effective as a rightly timed pause."
- Mark Twain
Full result tables for the different questions can also be downloaded directly from Google here, here and here.
"Do you agree that Quebec should become sovereign after having made a formal offer to Canada for a new economic and political partnership within the scope of the bill respecting the future of Quebec and of the agreement signed on June 12, 1995?"
- 1995 Quebec Referendum
THE ANALYSIS Just by changing the words used you can have a dramatic shift in results. Use the word "agree" instead of the word "should" and all of a sudden the yes side jumps a few points. Use "update" instead of "change" and you've shifted an election.
And that, ultimately, is the problem with referenda. How democratic and fair is it if changing two words can swing the outcome by 15 points?
Referenda turn complicated questions into false dichotomies that are manipulable by the political class in a way that your representatives are not. After all, no matter how many times - and no matter how many different ways - you ask the House of Commons the same question you're going to get the same answer. And that's the way it should be.
Democracy is not synonymous with voting. The Soviet Union had elections. Saddam Hussein's Iraq had referenda. Democracy requires an educated populace, fundamental freedoms and the rule of law.
Democracy also requires thoughtful deliberation, compromising between multiple points of view and mechanisms through which popular will can be accurately reflected. A referendum fails to deliver on any of these.
It has become fashionable for the Conservatives to say Canadians deserve a referendum on this question. But Canadians deserve better than a referendum on this. They deserve the best thinking of our elected representatives.
Let's do this the right way.

This site is the creation of Corey Hogan a political pundit who is also one of three guys who make up The Strategists, a podcast about Canadian politics even more beloved than Saddam Hussein was in 2002. He'd love to hear what you think about all this: tweet him @coreyhogan. This effort is not affiliated with any political party, provincial or federal.
  1. I guess that's why it's not called Google Voter Surveys.
  2. Source: Elections in Europe by Dieter Nohlen, via Wikipedia. Impressively, in 1966 turnout was 100% with Communists and allies getting 100% of the seats.
  3. Source: BBC. In 2002 apparently every single Iraqi voter voted in a referendum on Saddam Hussein's presidency - giving him 100% backing.