Fixing things not broken: The long-form census

The  furore caused by the Canadian government’s recent announcement -- that it would replace the mandatory long-form census questionnaire that is sent to 20% of the Canadian population, with the National Household Survey that will be sent to 33% of the population, but without legal obligation to complete it = is understandable. The problems that will result from this decision have alarmed the many agencies, municipalities, academics, school boards, charitable organizations and industry analysts who depend on the detailed data from this census for policy making, social and economic research and planning. Issues of compliance, introduction of self-selection bias, and the inability to compare results reliably over time – are very real and should give the government pause.

Yet, Federal Industry Minister Tony Clement insists the government has no intention of changing its decision, which was in response to complaints from citizens that the mandatory long form was invasive and coercive. However,  the Office of the Privacy Commissioner apparently received only three complaints about any aspect of the census in the past decade. Yes, those who have filled in the form have found it long, some of the questions may seem invasive, and being threatened with a jail term if one refuses to fill in the census is rather extreme.  However, given the vital importance of the data and its usefulness to the economic and social lives of Canadians, can the issue of invasiveness be better addressed through education (by, for example, co-opting the help of those chosen to complete the long-form by illustrating how important their help is to the process, and by explaining that the data is only available in the aggregate, that no individual data is revealed to those who use the data)? Can compliance be encouraged without extreme legal action (for example, by providing incentives to those who are required to fill in the form)?

The Government appears also to be pulling rank in going ahead with its decision. In response to Chief Statistician Munir Sheikh’s resignation over this issue, Tony Clement pointed out that Statistics Canada is not an independent agency.  This is true, Statistics Canada is an agency of the government as defined by the Statistics Act.    However, as reported on the Royal Statistical Society website, the government should take heed of the exhortation of David Green of the University of British Columbia, that Statistics Canada be allowed to operate at arm’s-length from political interference, as the Bank of Canada has for the past decade. "The same should be true of the national statistical agency. If statistical collection changes with the ideological whims of the government, the very basis of government decision-making, transparency and trust is shattered".

Statistics Canada’s purpose is to provide statistical information and analysis about Canada’s economic and social structure, and to promote sound statistical standards and practice.  It enjoys an excellent reputation in the conduct of its mandate.