Reviving the census debate

The Liberals and the NDP have said they want to bring back the long-form census the Conservatives killed in 2010. Could it become an election issue?

Thestar.com
Sept. 12, 2015
By Donovan Vincent

Canadian researchers Daniel Wilson and David Macdonald say they are facing enormous stumbling blocks due to the federal government’s elimination of the mandatory long-form census in 2010.

The pair, doing work for the Canadian Centre for Policy Alternatives (CCPA), a non-partisan research body that focuses on social, economic and environmental issues, is struggling to reconcile trends they’re now seeing in child poverty rates among native children.

The problem: they’re comparing data between the 2006 mandatory long-form census and the new - optional - long-form National Household Survey (NHS) that the federal government introduced in 2011.

Because the data from 2006 and 2011 came from two different processes, the researchers say they can’t tell if the latest trends they’re seeing are real or due to the fact so many fewer people filled in the optional long form in 2011.

“The practical challenge with working with the NHS is doubt - doubt that what you’ve found isn’t what’s actually happening in the world, but rather is a statistical artifact,” says Macdonald, who is also an economist.

Researchers, public policy advocates, statisticians, business groups, economists - and the Liberal and NDP parties - continue to call for the mandatory long-form questionnaire to be brought back, arguing that important statistical data is getting lost.

In a package of recently proposed reforms on transparency, the Liberals are promising to immediately restore the mandatory long form if they form government in the Oct. 19 federal election.

And Jean Ong, a spokesperson for the NDP, said in a statement that the party has long advocated for the restoration of the long-form census and continues to do so.

The lost data has massive implications for public policy decisions, business planning and a host of other areas, proponents of the mandatory long survey say.

Yet so far, the census hasn’t been in the spotlight on the campaign trail. But could it become an election issue?

Paul Jacobson, a Toronto economics consultant who relies heavily on census data for his work, believes it should. He says business planning is being seriously harmed by the new census data collection system.

“All the money in the world given to business surveyors could not replace the (mandatory) long form, period. You need a mandatory survey to get the quality of data you need to make good comparisons in small areas. That’s how you do business planning,” Jacobson says.

Stephen Toope, president of the Federation for the Humanities and Social Sciences, a national public policy advocate for Canada’s scholars, students and practitioners in the humanities and social sciences, says the “essence of the concern” about not having the mandatory long-form census is the impact on public policy.

“Thinking about questions around immigration, social service, children’s health and what kind of investments need to be made and where they need to be made - if we don’t know who is where, it’s very difficult to make informed policy decisions,” Toope says.

Researchers Wilson and Macdonald had found, based on the 2006 mandatory long form, that indigenous children in this country are more than two and a half times more likely to live in poverty than non-indigenous children.

However, in western cities where there are large indigenous populations, like Regina and Saskatoon, indigenous child poverty rates have fallen, in some cases dramatically, based on the new census data.

But for other disadvantaged groups in Canada, including some visible minorities, child poverty rates have not declined, nor have they declined in these same cities for the broader population, Macdonald said.

Bruce Campbell, the executive director of the CCPA, says this is but one example of the continuing challenges researchers are facing in obtaining reliable data based on the new census.

The response rate from Canadians to the mandatory long-form survey had been 93.5 per cent, but that dropped to 67 per cent in 2011 when the optional one was introduced. In addition, census data on more than 1,000 Canadian communities had to be withheld because response rates for those areas in 2011 weren’t high enough to draw proper conclusions.

“If there are these kinds of gaps (in data) it’s not showing the kinds of trends you want to track,’’ Campbell says.

Compounding the problem is that when it comes to the voluntary census, groups that don’t tend to respond are poor, and the marginalized, including individuals on native reserves, Campbell says.

“If our data shows First Nations child poverty is falling, is that because not enough people are responding to the questionnaire?” Campbell wonders.

Since 2011, selected Canadian households have only been required to answer a short eight-question census form, which asks for details such as income levels, citizenship, employment and ethnicity - down from the 53 questions respondents were required to answer in 2006 and earlier on the long form.

The mandatory short form asks questions including how many people reside in the home, their names, the citizenship and immigration status of residents, the languages spoken, custody arrangements for children in the home and marital relationships.

In explaining the Conservative government’s decision to kill the mandatory long form in 2010, Tony Clement, the federal industry minister at the time, told a standing committee in Ottawa that his government’s reason was simple.

“We do not believe it is appropriate to compel Canadians to divulge extensive private and personal information,’’ Clement said at the time, referring to questions such as: How many sick days did you take last year? Were you paid for those? What were your total payments for your primary dwelling last year?

Clement said the government sought options with Statistics Canada on how to implement a reliable voluntary survey.

“This reasoned and responsible approach is about finding a better balance between collecting necessary data and protecting the privacy rights of Canadians,’’ Clement said at the time, adding that on the advice of StatsCan, which recognized that the sample size would decrease with the long form being voluntary, his government decided to send the new optional long National Household Survey (NHS) to 4.5 million Canadian households, almost double the sample size from 2006.

Since 2011, private member’s bills from the Liberals and NDP aimed at bringing back the mandatory long questionnaire have been voted down in the House of Commons.

One of the reasons the Conservatives did away with the long questionnaire in 2010 was threatened jail terms for failing to fill out the mandatory long form. In June, a Conservative private member’s bill that sought to do away with the threatened jail terms for failing to fill out the mandatory short-form census made it to third reading, but died on the order paper.

Lost census data

Proponents of bringing back the long-form census cite various negative impacts of the lost data, including:

Labour force planning: Census data on labour forces can be useful for tracking occupations in specific locations, providing details on the attractiveness for specific investment in given areas and the “relative sophistication’’ of industrial sectors, says Toronto consultant Paul Jacobson.

But that’s finer detail than can be obtained from the voluntary long-form system, he adds.

Tracking income inequality: Census data can be used by social policy groups and housing advocates to track neighbourhoods and communities where growing poverty might be a concern. “We can document trends. Looking at the last 20 or 30 years, especially among the top 5 per cent, top 1 per cent of income earners, there’s been a disproportionate increase in their income compared to the bottom 90 per cent of the population,’’ says Bruce Campbell, the executive director of the Canadian Centre for Policy Alternatives. Such census data has huge implications for public policy, he says.

Aboriginal education: Social policy advocates are concerned that marginalized groups, including indigenous Canadians are less likely to fill out the optional long form. “If we don’t know where aboriginal people are living - many don’t live on reserves - how are we going to figure out how to serve aboriginal populations (through schooling)?’’ says Stephen Toope, president of the Federation for the Humanities and Social Sciences.