Toronto professors explain why Canada’s long form census matters

Canada’s long-form census will be back for 2016, with packages ready to be mailed out on May 2 to approximately 2.9 million households, much to the relief of academic researchers, policy analysts and healthcare administrators everywhere.

“We need good, reliable data,” said Minister of Innovation, Science and Economic Development Navdeep Bains on announcing of the census’ return.

“We know the history of the past government and they very much focused on ideology,” said Bains. “We’re focused on sound, evidence-based policies. We want to make sure we’re driving good policies based on good evidence and quality data.”

Michael Carter, co-director of the University of Toronto’s Centre for Research and Healthcare Engineering, told U of T News, “We are very grateful, so pleased to know it is back,” noting the value of information provided by the long-form census for determining greatest need in building Local Health Integration Networks in Ontario.

“One of the major concerns with health care is that it is relatively easy to measure utilization but how do we measure the true demand for it? The long-form census is a valuable asset for that,” said Carter.

Timothy Chan, an associate professor in U of T’s department of Mechanical and Industrial Engineering, used long-form census data detailing population density and demographic data to determine where to place defibrillators in public places based on populations most at risk for cardiac arrest.

“The more granular the data, the better,” said Chan, noting that reliable census data is “extremely critical” in determining practical aspects of health policy.

In 2010, the 61-page long-form census was replaced by a voluntary National Household Survey, which cost $22 million more than the long-form census to develop and administer while also delivering poorer quality results.

The response rate for the National Household Survey in 2011 was 69.3%, compared with a 93.5% response rate for the long-form census. Response rates were lower in smaller and mid-sized cities, precisely the places where reliable data is essential to administering social policy.

“The purpose of a census is to provide data for the smallest regions and the smallest populations and with the lower response rate,” said Canada’s Chief Statistician Wayne Smith to McLean’s. “What the NHS showed clearly is you’re going to lose that.”

It was in 2011 that Statistics Canada warned that information collected for smaller communities was spotty enough that it wasn’t worth releasing to the public, mainly because of data bias.

People more likely to fill out a voluntary census form are the wealthy and middle class, meaning results would be skewed.

In July, ScotiaBank released a report attempting to gauge the effect of a potential rate cut by the Bank of Canada on the Canadian economy.

“It could be rather ill-advised for the BoC to cut based upon stale data,” wrote ScotiaBank vice-president and Head of Capital Markets Derek Holt. “By StatsCan’s own past admissions, much of the trade data on the resources and particularly the energy side of the picture is inferred because it is not available on a timely basis for the first pass before completed industry responses are incorporated into revisions.”

Holt was acknowledging a fact widely known by financial analysts and anyone else trying to determining credible predictive policy in a country whose main statistical agency, Statistics Canada, had been disemboweled over the course of a decade by a government that valued ideology over results or evidence or practical outcomes.

Alarmingly, Holt’s reticence to advise on monetary policy was an admission that administrators have been struggling through a fog of imprecise numbers while trying to plot a course of action that has a real effect on the lives of Canadians.

The ideological rationale for the eliminating the long-form census was privacy. But pursuing that ideology harmed both the Canadian government’s ability to craft policy and the country’s entrepreneurial and business classes.

Alan Walks, associate professor of Geography at the University of Toronto told U of T News, “It’s kind of ironic because the previous government was arguing against having the long form on privacy grounds, but municipalities, provinces, businesses, civil society organizations, universities said the data was useful, and there were very few reported cases of people complaining because of privacy issues. It didn’t make any sense; it was actually harming Canadian business, harming the competitive aspect of businesses and entrepreneurs not to have this data.”

On straight privacy grounds, too, if you’ve got a loyalty card or participate in any points program in Canada, not to mention if you’re on Facebook or use Über or ever Google anything, you’ve already given away far more valuable information about yourself, essentially in exchange for a biscuit, than for a government-administered census, without the added benefit of helping Canada forge credible economic and social policy.