The federal government should plead temporary insanity and reinstate the long-form census. In 2010, Prime Minister Stephen Harper asked Statistics sa国际传媒 to replace the mandatory long-form census with the voluntary National Household Survey, saying Canadians were concerned about privacy, although there didn鈥檛 seem to be much of an outcry in that regard.
However, a storm of protest followed the decision to make compliance voluntary. It came from municipalities, economists, historians, genealogists, cultural and religious groups, and others who depend on the information. They said that without mandatory compliance, the information collected would be much less reliable.
Statistics sa国际传媒, it appears, agrees with the critics. Last week, it released the first pack of information from the 2011 survey with this statement: 鈥淭he NHS estimates are derived from a voluntary survey and are therefore subject to potentially higher non-response error than those derived from the 2006 census long form.鈥�
The agency cautions against comparing information from the 2011 survey with the 2006 long-form census data.
All households receive the short-form census, which asks basic questions. It is still mandatory. In 2006, about a fifth of Canadian households received the long-form census, which harvested detailed social, economic and cultural information. The response rate was 96.6 per cent.
About a third of Canadian households received the NHS in 2011; 68.6 per cent completed it fully.
Eric Sager, a University of Victoria historian who has used census data extensively, said that raises serious concerns about the reliability of data. It鈥檚 not just that the response rate is lower, he said, it鈥檚 the kinds of people under-represented, particularly minority groups, First Nations and recent immigrants who might not be inclined to complete the form.
鈥淗istorians are concerned about the legacy we leave to our descendants in the distant future,鈥� Sager said. 鈥淭he absence of the long form is going to be serious.鈥�
The issue is not just about preserving information about the past. Planners and policy-makers need reliable information as they assess such needs as transportation networks, schools and hospitals. Sager said the 2011 survey data are not reliable enough for that sort of planning.
Paul Bramadat, director of UVic鈥檚 Centre for Studies in Religion and Society, agrees.
鈥淚 think it creates major problems in any efforts to compare previous years,鈥� he said. 鈥淪witching to a voluntary system has made it difficult to get a clear sense of what鈥檚 going on.鈥�
He said last week鈥檚 data present much the same general picture as the 2006 census. 鈥淚t鈥檚 just that some of the particular numbers are puzzling,鈥� he said.
The Harper government鈥檚 concerns over privacy are groundless.
鈥淚 have said time and again, emphatically, there is no invasion of privacy,鈥� Sager said. 鈥淚nvasion of privacy occurs when personal and private information is made available to the public. That does not happen to census information.
鈥淚t goes into computers. It cannot be seen or read by anyone outside. It can鈥檛 even be read, in most cases, by Statistics sa国际传媒 employees. It appears only in aggregated form, with no personal identification information. There is no risk to privacy whatsoever.鈥�
Bramadat said Statistics sa国际传媒 is known internationally for its high standards of data collection, a reputation threatened by the switch to a voluntary survey.
He regards the 2011 survey as a blip.
鈥淢y hope, of course, the hope of many academics, is that the government will go back and make the data collection more rigorous and consistent,鈥� he said.
Information is gold, but only if it鈥檚 reliable. The federal government should chalk up the survey as a failed experiment and let Statistics sa国际传媒 return to the mandatory long form to collect data in a consistent and scientific manner.
