By David Bush
On Tuesday, October 10 NewsTalk 1010 published a poll showing relatively tepid support in Toronto for the Liberal’s plan to raise the minimum wage in Ontario to $15/hr. The poll claims to show 61% of Toronto residents don’t want the province to move ahead with the $15/hour minimum wage rollout as planned. It also says that 44% want a longer phase-in, while 17% want to abandon the $15 minimum wage altogether.
This would suggest that the Liberal plan to raise the minimum wage $15 does not have strong support. This is especially shocking considering the last poll of Toronto residents, done by Forum Research in April 2017, showed overwhelming support for a $15 minimum wage. In the Forum poll 70 percent of residents approved of raising the minimum wage to $15, while only 25 percent disapproved.
On the face of it, the two polls would suggest there has been a major shift in support for a $15 minimum wage. But a closer look at the poll commissioned by NewsTalk 1010 shows it is highly skewed and unreliable.
Selection bias
The poll, conducted by DART Insight and communication between September 16 – 19, surveyed 814 people online through the social media site LinkedIn. Unlike phone surveys the self-selection bias in conducting a poll through social media is much higher. Among the four major social media platforms, Instagram, Twitter, Facebook and LinkedIn, LinkedIn has a much wealthier user demographic. The latest PEW survey show that 45 percent of adults with a household income at or above $75,000 have a LinkedIn account, far outstripping the user rate from any other income bracket. Facebook and Instagram show other lower income brackets have a larger percentage of user accounts than the over $75,000 bracket. LinkedIn skews towards wealthier users, which shouldn’t surprise anyone, as it is a social networking platform designed for professional advancement.
The use of LinkedIn may explain why well over a quarter of the participants in the DART poll had a household income over $100,000. While the results were weighted this perhaps explains why the margin of error in the poll is +/- 3.9 percent, which is a higher margin of error than typical for a poll its size. It is hard to say with certainty without seeing the full data.
Leading question
Beyond the problems of self-selection bias, there is a much bigger problem in the poll, and that is the nature of the question. If one wanted to know the level of support Toronto residents have of the $15 minimum wage and the Wynne government’s planned implementation one would aim to ask a clear and direct question, something like this:
“Do you favour the Ontario government’s plan to phase in a $15 minimum wage by January 1, 2019?”
This is simple, direct and does not aim to sway the opinion, but gage it. The Forum poll in April asked the question:
“Currently, the minimum wage in Ontario is $11.40 per hour. Would you
approve or disapprove of raising the minimum wage to $15.00 in the
province?”
Simple and to the point. The DART poll’s question was the exact opposite:
“Ontario’s Liberal government is raising the minimum wage by 32% to $15 per hour over one-and-a-half years with some breaks to business but the timetable is firm. Neighborhood businesses of all sizes say they need more time to absorb the increase. Also, the Government’s own independent Financial Accountability Office (FAO) says it could cost upwards of 50,000 jobs. Do you think local businesses should get more time to phase the increase, or should the government just move ahead, or should the government just cancel the entire thing and let the marketplace figure it out on its own?”
This is a textbook case of a leading question. Rather than gauging support, the question aims shape the results. The question assumes that all businesses think the same way, when other polls and prominent voices in the small business community have stated they support the minimum wage increase. A June, 2017 Campaign Research poll shows that a majority (54 percent) of small and medium enterprise (SME) owners approved of a $15 minimum wage, while only 30 percent who found $15 too high.
So while some SME business owners have expressed vocal opposition to the $15 minimum wage, those voices it turns out are in a minority. Including the vague wording of “neighborhood businesses of all sizes say they need more time to absorb the increase”, not only reflects the bias and poor research skills of the pollsters, it constructs a very specific and tainted framework for the question. Instead of this question being about the minimum wage, it is about the minimum wage as it relates to the supposed interests of small business.
The FAO study
A second major problem in this question is the use of the Financial Accountability Office’s report on the $15 minimum wage. The FAO’s study has come under some serious questioning by other economists over their use of imprecise language and scholarly estimates in coming up with 50,000 fewer jobs being created over the long-term. Jordan Brennan, an economists at Unifor noted about the misrepresentation of the FAO study:
“I doubt the media will note this, and part of the problem flows from language selection, but there is a difference between an existing worker being laid off and the rate of (future) job creation slowing down. In in the former scenario an actual person is made materially worse off, while in the latter situation, a hypothetical worker—someone who is not presently employed, but who may seek work in the future—is not able to find a job.”
The inclusion of the job loss numbers from the FAO study is a clear case of bias. The FAO study also concluded that 1.6 million workers – 22 percent of the labour market – will see a positive direct impact, while workers earning $15 to $19 will likely see an indirect pay increase. The pollster chose instead to include the misleading fact about potential job deceleration over an actual concrete statistic in the same study.
Looking at all other other different questions in the same poll — mayoral preference (only 19 percent of Toronto resident want Drake as mayor), the role of the police, Marijuana legalization, publicly funded daycare etc.— no other question was as lengthy, inserted study findings, or constructed scenarios of possible outcomes.
Reject the push poll
This was a push poll aimed at showing the weakening of the support for a $15 minimum wage. What is surprising is the strength of support despite the obvious bias in the question. Only 17 percent of respondents favoured cancelling the initiative, while a full 83 percent wanted to see a $15 minimum wage. Two in five said they wanted to see the 18-month phase-in of $15 go forward despite being confronted with a question designed to get them to choose a longer phase-in or oppose outright.
Bias and misleading polls tilted towards employers are nothing new. As the fear campaign by employers ramps up and we near the election expect more polls aimed at stoking fear and sowing doubt about raising the minimum wage and strengthening labour law. The media will not do our job for us.