Household survey or long-form census? Tories to choose (again) for 2016

TORONTO and OTTAWA – Just when you thought the national fulmination over the National Household Survey was over, Canadians will choose once again how to count themselves.

Or rather, their federal representatives will.

Statistics Canada, having released the final tranche of data from its inaugural National Household Survey, is preparing recommendations for Industry Minister James Moore regarding what to do for the 2016 census.

The federal Conservatives scrapped the mandatory long-form census in 2010, replacing it with a voluntary National Household Survey. But the Tories can choose to reverse that decision the next time around.

“What remains to be decided is what would be the content of the census itself on a mandatory basis, what would be the content or the mix of the voluntary questions … if we would repeat the NHS or not. Those are decisions the government has to make,” said Marc Hamel, Director General of Statistics Canada’s Census Management Office.

“We will put forward our recommendations on the models that should be adopted, and we will wait for a decision from the government on how to proceed.”

National Household Survey results lack the accuracy of a mandatory long-form census because not everyone responds to an optional survey, and some groups are more likely to respond than others.

We have some idea of who’s underrepresented – the very poor and the very rich, for example. But “we don’t know what we don’t know,” as Centre for Policy Alternatives economist Armine Yalnizyan puts it.

Hamel insists the accuracy’s just as good on the national or provincial level. But he admits it’s flawed for smaller or more rural communities.

“As you start looking at some of these results for smaller populations, the smaller areas, you might see a little bit more volatility in the information. So we are cautioning users,” he said.

Ironically, these are the ones that need the data most: The census is the only source of information every five years that looks at small towns, small neighbourhoods, small groups of the population,” Hamel said.

Watch: Can you trust the National Household Survey?

Neither Hamel nor Moore’s office would say when the Minister needs to decide.

“’As always, we continue to look at ways to improve the next census cycle while protecting Canadians’ privacy,” spokesperson Sebastien Gariepy wrote in an email.

In the meantime, groups reliant on census data are trying to decide whether they can trust this survey’s numbers.

The city of Vancouver “has concerns” the data will hamper the city’s ability to identify trends “and target services where they are needed most,” a spokesperson wrote in an email.

Toronto has already said it won’t use household survey data for trendlines at all.

Ontario is still assessing the survey data, said Finance Ministry spokesman Scott Blodgett.

Ontario’s Education Ministry is reviewing survey data “for several school-aged variables,” spokesperson Derek Luk wrote in an email, trying to decide “how reliable and usable the data would be for the ministry’s calculation of trend lines and conduct of other analysis at the neighbourhood level.”

Yalnizyan doubts this government will have a change of heart and revert to a long-form census five years from now.

But she contends a lack of good data will compound problems many Canadian communities are already wrestling with: integrating immigrants and providing affordable housing among them.