* Any views expressed in this opinion piece are those of the author and not of Thomson Reuters Foundation.
Canada dropped mandatory questions on poverty from its national census, making it harder to measure progress on anti-poverty programmes. The change illustrates why experts want the UN to include better quality statistics in its post-2015 development goals
TORONTO (Thomson Reuters Foundation) – Like many of her colleagues across Canada, Christina Maes Nino, an urban policy analyst in Winnipeg, Manitoba, is in a fog when it comes to measuring poverty.
As she works on a strategy to improve housing conditions in the city's impoverished Centennial neighbourhood, where housing is largely rundown and aging, she is missing key information about one of Canada’s poorest communities. She doesn’t know whether the last housing strategy worked.
While Canada is one of the world’s richest countries, a recent survey estimated that 15 percent of its people live in poverty – a problem that anti-poverty activists and planners say is very hard to address without adequate data to measure whether their development programmes are working. In fact, compiling reliable statistics is a global problem that plagues development efforts, one that a panel of experts highlighted recently when it recommended that the United Nations include better data collection as a priority in its post-2015 development goals.
In Canada, a change in the way it conducted the national census in 2011 has led to huge data gaps that have blindsided urban planners and weakened Canada’s ability to respond to what Miloon Kothari, the former UN Special Rapporteur on the Right to Adequate Housing, described in 2007 as a “national housing crisis.”
Planners across the country say they are concerned that without reliable numbers, the needs of the poor will not be considered in policy decisions and that it will be difficult to promote effective investments in poor communities. To make matters worse, policymakers don’t have the data they need to measure how effective previous policies have been.
“If we don’t have good data about how that neighbourhood changes over time, then we don’t know if anything the [housing] plan is putting down as a goal is being achieved,” said Maes Nino.
Maes Nino’s dilemma began in 2010, when the Canadian government acted on concerns about privacy and scrapped mandatory questions related to poverty from its 2011 national census. Statistics Canada, which was once recognized internationally for its data expertise, kept a mandatory short questionnaire of 10 questions in the census. But the previously mandatory long-form questionnaire, which included questions on income, housing and employment, was replaced with a voluntary survey called the National Household Survey (NHS).
The decision was controversial. Hundreds of organisations, ranging from church groups to businesses, criticised the move, saying a voluntary survey would provide less accurate information about the population. The government’s chief statistician resigned.
Now results from the 2011 survey are out. While it indicates that about 4.8 million Canadians, or 15 per cent of the population, are poor, the published data contains significant gaps. In some parts of Canada, so few people responded to the survey that Statistics Canada, out of concern for data reliability, did not publish numbers for 1,128 municipalities and several urban neighbourhoods.
“To just frame it as a policy matter, I think, would be incorrect. This data is important in order for people’s rights to be recognised and fulfilled,” said Leilani Farha, the executive director of a non-profit advocacy group Canada Without Poverty (CWP).
On its website, CWP criticises the Canadian government for failing to act on recommendations related to poverty reduction made during the UN Human Rights Council’s 2013 review of Canada.
In addition to data gaps, policy analysts and planners are concerned about data reliability.
“We don’t know if we’re adequately capturing the poverty story,” said Armine Yalnizyan, the senior economist at the Canadian Centre for Policy Alternatives. She noted that research shows that poor people tend to respond to voluntary surveys at low rates.
While the response rate matters less for data at the national level, data at the local level can be easily skewed if a survey isn’t filled out by a group that is representative of a neighbourhood, said Richard Shearmur, a professor at the McGill School of Urban Planning in Montreal, Quebec.
Planners say there are limited alternate data sources, and none offer the same level of detail as the census. This limits the ability of planners to respond to the needs of poor communities.
Saint John, New Brunswick, lacks data for 13 neighbourhoods, including three of the city’s poorest, which makes it challenging to set policy.
“There’s a large amount of sub-standard housing in Saint John. We know that just by visual inspection, but ... there’s nothing like the rigour of data to help influence policy change,” said Randall Hatfield, the executive director of the local Human Development Council.
The situation is made worse by Statistics Canada’s warning against comparing NHS data for income with previous censuses because the methodology changed. This makes it challenging for urban planners to track changes in poverty over time – a process that helps them determine which policies are working and which ones are not, said Damaris Rose, a professor of urban studies at Quebec’s National Institute for Scientific Research.
With or without the data, the planning continues.
“We’re still going to do everything we can to reach a diverse proportion of people. But we can’t say for sure that we’ve done a good job of that,” Maes Nino said.
Alia Dharssi is a fellow in global journalism at the Munk School of Global Affairs at the University of Toronto and is interning at the Thomson Reuters Foundation
Our Standards: The Thomson Reuters Trust Principles.