Laurier professor weighs in on census debate

When Tony Clement, the federal minister of industry, announced that the mandatory long-form portion of the census would be replaced with a voluntary survey starting next year, he may have been expecting this usually uncontroversial topic to go unnoticed at a time when parliament is not sitting and Canadians are generally not focused on political events.

However, over the past month the issue of the long-form census has generated a great deal of controversy and has served to unite a wide variety of organizations against the government’s position.

“What is remarkable about the census debate is that the opposition spans Catholics and Evangelicals, labour unions and bankers, provincial and municipal leaders,” said Laurier political science professor Andrea Perrella. “I have never seen so many disparate groups unified.”

These organizations rely on accurate information about the Canadian population in order to carry out their mandates. A great number of such groups believe that a voluntary survey cannot provide them with data that is as accurate as a mandatory census.

“They cannot compare [the new survey] to previous years. If they change it back, they cannot compare it to the next census. There’s no scientist on the planet who would accept a series where the data points are based on different metrics,” explained Perrella. “The census has an advantage because of the large sample over all communities. A voluntary survey can give general information, but that doesn’t guide policy. More specific information guides policy.”

The government, however, has defended its position by saying that the mandatory long-form census is an unwarranted intrusion into the privacy of Canadians. One question in particular regarding the number of bedrooms in the house is being exemplified as being of an intrusive nature.

However, media inquiries revealed that only three complaints were made to the Privacy Commissioner in relation to the last census in 2006.

“Privacy concerns are a red herring. We [members of the public] release private information every day,” said Perrella. “There are easier and more lucrative ways to violate privacy. There’s a far greater risk in everyday activities.”

He added that the data released to the public is aggregated and is not broken down on an individual level.

In previous censuses, every Canadian household received the short form, which asks eight basic questions about the occupants of the household. In addition, one in five Canadian households received the long form, asking a variety of more detailed questions.

The questions on the short form include the names, ages, marital status and first language of the members of the household, while the longer voluntary survey has additional questions relating to daily activities, health problems, place of birth, citizenship and immigration, languages understood and used in the home, ethnic origin, religion, education, employment history and income.

The long-form questions will now be asked as part of a voluntary survey rather than a mandatory census. In order to ensure that Statistics Canada receives a sufficient number of responses, the survey will be sent to one-third of Canadian households rather than one-fifth as was done with the mandatory long-form.

Printing and processing these additional forms and advertising to encourage people to respond are expected to result in extra costs of up to $40 million dollars. Because a survey is not mandatory and members of certain groups are more likely than others to voluntarily participate, questions arise as to whether the data may be biased and thus not present a fully accurate representation of society.

Perrella also expressed concern that a lack of accurate data may leave any future government policy more open to opposition.

“Less information makes it more difficult to govern well,” he said. “I don’t see how it would benefit the formulation of sound, intelligent policy.”

On July 21, the head of Statistics Canada, Munir Sheikh, resigned. In a statement released at the time, Sheikh partially explained the reasons for his resignation.

“I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion. This relates to the question of whether a voluntary survey can become a substitute for a mandatory census. It cannot.”

When Sheikh appeared before a committee of the House of Commons later in July, he stated that he resigned because the public perception was that Statistics Canada had assured the government that a voluntary survey could be as accurate as a mandatory census. Since it is generally accepted that a voluntary survey cannot be as accurate as a mandatory census, Sheikh decided to resign in order to protect the agency’s reputation and credibility.

Despite the opposition to the elimination of the long form census, there is little evidence that the government is seriously considering a reversal of policy. The government appears determined to replace the mandatory long-form census with a voluntary survey.

“No idea sees the light of day without PMO [Prime Minister’s Office] approval. The PMO is an unelected and unaccountable institution. It is not answerable to the civil service, parliament or voters,” Perrella concluded.

Opinion Editor Eric Merkley weighs in on the long-form census debate.