Monday, October 4, 2010

The Census Long Form and Vital Signs

On Wednesday September 29th, the House of Commons passed a motion calling on the federal government to reinstate the mandatory long-form census. This is just the most recent of many efforts by businesses, academics, community organizations, individual citizens and politicians from all parties and levels of government to convince the federal government to change its course. Community Foundations of Canada has been part of this effort and in July wrote a letter to Minister Tony Clement.

The loss of the mandatory census long form represents a disaster for Canada's statistical infrastructure. The resignation of Canada's Chief Statistician shows the seriousness of this issue. But this bad political decision is reversible. It appears not to be possible to reinstate the long form for the 2011 census (although some argue this still can be done by postponing the census to the second half of 2011), but the long form can certainly be reinstated for the 2016 census.

It is important that Canadians make their views known on this issue in as many ways as possible. One way is by signing the national petition to reinstate the long form. Another way is to contact your local MP and let them know that you support the reinstatement of the long form census.

The demise of the mandatory long form for the 2011 census is a blow to the Vital Signs project organized by Community Foundations of Canada, as it is for other projects that rely on this data such as the Canadian Index of Wellbeing and Federation of Canadian Municipalities' Quality of Life Reporting SystemThe census provides high-quality, reliable data for Canadian communities of all sizes. It is the most important data source used by local community foundations to compile their Vital Signs reports.

As part of the research support for the Vital Signs project, the Centre for the Study of Living Standards developed and maintains a comprehensive data base for Vital Signs communities containing about 70 indicators in ten issues areas. The census is a source 18 of these indicators, and is more important than other data sources such as the Labour Force Survey (LFS), the Survey of Labour and Income Dynamics (SLID), and the Canadian Community Health Survey (CCSH). The census is particularly important for small Vital Signs communities, such as Wolfville, where the limited sample size of national surveys such as the LFS means that estimates for these communities are just not possible.

Even for the CMAs where estimates from other household surveys such as the LFS, SLID, and CCHS are often available, these estimates are much less reliable than census data because the census long form has a much greater sample size (20 per cent of the population). More important, information from the census long form is used to benchmark and weight the different groups in other household surveys. The demise of the long form means that the quality of the estimates in these surveys will deteriorate in the future because these benchmarks and weights cannot be updated.

Looking beyond the needs of the Vital Signs project, the loss of the census long form means that the community profiles posted on the Statistics Canada website, which are freely accessible to all Canadians, cannot be reliably updated.

These profiles, which unfortunately are much less known and used than they should be, provide estimates for such variables as population, gender, age group, characteristics of dwellings, educational attainment and field of study, marital status, family and household characteristics, earnings and income, mother language, knowledge of official languages, labour market status, employment by industry and occupation, mobility status, Aboriginal and visible minority status, immigration and citizenship status, place of work, and mode of transport to work, for all cities, towns, municipalities, and reserves in all provinces and territories in Canada. Data for literally tens of thousands of communities are available.

For example, if one want to know the median income in 2006 of female lone-parent families in the small town of Vulcan, Alberta (population 1,940), one can find it on the community profiles website with several keystrokes ($34,614). By replacing the mandatory census long form, which was answered by a random sample of 20 per cent of Canadians, by the non-mandatory National Household Survey, the results for the 2011 census will be much less reliable for the construction of community profiles than was the case in previous censuses, if they can be used at all.

Again,it is important that Canadians tell the government that we want reliable data to be available to policy makers, the not for profit sector, researchers, advocates and the public. Please sign the national petition to reinstate the long form http://www.savestatcan.ca/ and contact your local MP and let them know that you support the reinstatement of the census.

Andrew Sharpe
Executive Director
Centre for the Study of Living Standards
Ottawa, Ontario
andrew.sharpe@csls.ca

No comments:

Post a Comment