Statistics Canada has removed questions about unpaid work from its replacement questionnaire for the contentious long-form census - an omission critics fear will lead to misdirected policy for seniors, women and children.
Industry Minister Tony Clement has repeatedly defended the government's decision to axe the mandatory long census by saying it will be replaced by a voluntary survey that asks the same questions. But the voluntary survey, called the National Household Survey, is not an exact replica.
It adds a few questions on commuting, subsidized housing, child support payments and religion, Statistics Canada officials said. It also deletes an entire section about what Canadians do when they're not working for money.
In the 2006 long census, Statistics Canada asked respondents how many hours a week they spent doing unpaid housework, yard work or home maintenance.
They also asked how much time they spent looking after their own children or other people's children, without pay - including helping kids with their homework and talking to teens about their problems.
And they asked how much unpaid time they dedicated to looking after seniors, including talking with them on the phone or helping them with medication.
"What's really going on here is that it's the unpaid care economy that's being removed from the detailed economic data that's being collected," said Kathleen Lahey, a law professor at Queen's University in Kingston, Ont.
"They've decided to erase the care economy."
The omission means there will no longer be a robust collection of data that helps everyone from policy makers to marketers plan for what people do when they're not at the office, she said.
"Almost every policy option touching on the family and community services is touched by that data," Prof. Lahey said.
When governments look at care of the elderly, child care, discrimination against women and productivity, the information collected in the mandatory long census was central and crucial, Prof. Lahey argues. "There is no other way to validly and accurately collect this information."
But Statistics Canada argues that the information is available through another agency survey known as the General Social Survey.
And the National Statistics Council of experts who advise the agency says getting rid of the questions on unpaid work was a good idea.
That's because there has never been a demand for information on unpaid work at the detailed, neighbourhood level that the long census provides, council chair Ian McKinnon said.
The long census questions were also too vague to be of any solid use, he added.
"It's not that it's an unimportant question," he added.
Research on unpaid work has revealed valuable information about productivity, human resources, aging and childcare, he said.
But the General Social Survey collects time-use information from about 18,000 people every five years, and provides a far more useful portrait than the long census questionnaire ever did, Mr. McKinnon said.
He concedes, however, that the General Social Survey and other Statistics Canada surveys will be less valuable in the future, because it will not be able to benchmark results against the now-defunct mandatory long census.
Discussions about deleting the unpaid work questions were in the works long before the federal Conservatives decided to axe the mandatory long census.
In a 2008 review, Statistics Canada asked for comments on the unpaid work questions, and received 54 responses. The review did not come to any striking conclusions.
It found that opinion was divided on the importance of the questions. Some said the information was central to understanding gender equity, economic divisions, volunteer work and social policy. Others said the questions were too broad and could be better answered in different surveys.
"This is not a proper basis for a decision that itself has such huge statistical and policy implications," Prof. Lahey said.