It’s election time and so – surprise, surprise – there is a lot of polling out there telling people who is going to vote for whom.  Tomorrow, I will talk about polling specific to post-secondary education and science, but today, I want to talk more generally about how polling frames election issues and how poor some of the framing seems to be going into this election.

In the pre-writ period, the most important question in polling is “what is the ballot question”: that is, what issues will the electorate be focussing on when they cast their ballot?  This tends to be a bigger deal in the development of opposition parties’ platforms than the development of the government’s. Opposition parties are forever trying to turn elections into referendums concerning the current government’s decisions on those ballot question issues. (Here’s a handy rule of thumb: incumbents want to frame elections as a choice, challengers want to frame it as a referendum). This year, there is a lot of confusion about what the ballot issue is because of divergent pollster approaches.

There are two methodology questions: whether to ask an open or a closed question and if the latter, what options respondents are allowed to pick.  The method that seems to predominate in Canada is a “closed” question, which typically produces the same kinds of issues election after election (“health care” if the economy is good and “the economy/jobs” if it isn’t).  This, from a market research company with the inventive name “Research,” is a pretty typical example: jobs and health both take the first spot with 19%.

But the thing about a closed approach is that different pollsters can reach different conclusions based on what options they provide their respondents.  Abacus Data has attracted attention over the last few months simply by adding “the cost of living” to the list of answers, and somehow finding that it comes tops.  Media are attracted by the novelty of the finding, and political parties take notice too; the NDP particularly seem to be making “cost of living” their signature issue for the campaign (and indeed, the NDP’s platform, released in June, is very specific in putting its post-secondary education commitments under “affordability” rather than, say, “building the economy.”) 

The idea that prices are rising faster than incomes is simply unsupported by data.  Sure, the CPI is an average and it doesn’t reflect everyone’s basket of goods, but even if it is true, as some argue, that certain societal groups might be facing above-average cost increases, then by definition there are a roughly equal-sized group of people for whom inflation is below­-average. 

I have written about this framing and why it could spell real trouble for higher education because it tends to lead to restricting institutional incomes, but I want to note that Abacus may be barking up the wrong tree.  That’s because Environics periodically asks a similar question (“what’s the most important problem facing Canada today”) but with an open methodology (that is, allowing people to answer how they want and then coding the results later) just 8% say “inflation/cost of living” (see page 6 of the linked document).

The easiest way to reconcile the Abacus and Environics questions is to assume that “cost of living” is, in fact, generalized anxiety about “keeping up with the Joneses” (which is indeed pretty hard – but it’s always been hard).  One piece of evidence in favour of this is that Abacus’s results don’t vary much by income bracket, rather than being prevalent among low-income Canadians or other specific groups facing price inflation higher than that measured by CPI (which is a thing) or those whose view of “cost of living” is coloured by the experience of looking for houses in Toronto or Vancouver.  One wonders what would have happened if Abacus had offered respondents choices such as “trying to get ahead” or “the rat race” as potential answers – my guess is they would have received very similar answers, but the issue then would not be framed in ways quite so harmful to higher education.

In fact, arguably what this is a continuation of something we’ve been seeing in the polls for a very long time – generalized middle-class anxiety.  Indeed, the current “cost increases are out of control” mania resembles nothing so much as the “squeezed middle class” mantra of 2014-15 – the one the Liberals rode to power four years ago.  It didn’t matter that, 2008-9 aside, the Canadian middle-class had been on an extended run of good fortune since 1998 or so, that average and median family incomes were at all-time highs, etc.  All this was shooed away as irrelevant because the middle class “felt” worse off.  Which is the main reason the Liberals can’t fight back against this largely fact-free line of attack: they used one almost exactly like it last time out, and their Finance Minister has been using the whole “feelings are more important than facts” line almost constantly since 2015. 

Long story short: the cost-of-living line is incorrect framing, and Abacus should know better, but it looks like politically we’re stuck with this nonsense for a while to come.