As our weekly maps are frequently based on data published by Statistics Canada, we have been eagerly awaiting the release of new datasets based on information collected during the 2011 national census. The Census of Canada, the flagship survey of one of the world’s most well-regarded statistical bureaus, has provided a wealth of demographic data on the nation’s population since 1871, helping shape social programs, infrastructure investment, economic policy and public interest research in communities across the country. The compulsory nature of the questionnaire (with all residents filling out basic information about themselves, and one in five households receiving a longer set of questions) has ensured that this statistical picture of Canadian life is both accurate and detailed, but with the Harper government’s controversial decision to eliminate the mandatory long census form in 2011, the quality of this valuable resource has been thrown into question.
The first wave of data from the National Household Survey, the voluntary questionnaire that replaced the long form, was released this week, containing a statistical breakdown of the ethnic and cultural backgrounds of Canadian residents as well as information about First Nations, Métis and Inuit communities. While many news outlets were eager to map the latest immigration and settlement trends, a number of experts took to the news and social media to point out one of the census’ most unsettling discoveries – not nearly enough people filled out the voluntary survey to provide an accurate statistical portrayal of many communities. This week’s map illustrates the varying proportion of households across the country that received the National Household Survey and did not respond to it.
In an attempt to compensate for the inevitable slice of the population who received the survey but chose not to answer it, forms were sent to approximately one-third of Canadian homes, more than the 20% of the population who received the long form in the past. As you can see from the map, non-response rates vary across the country, with fewer people choosing to respond in rural areas than in the urbanized regions along the border – while special efforts were made to enumerate remote communities and reserves in particular, the grand total of those who chose to fill out the NHS fell quite short of the mark set by previous censuses, and these respondents are unevenly dispersed across the country. In other words, the statistics released about areas coloured in green and blue might not be the most accurate representation of everyone who lives there. However, the most challenging insight from this map comes from all the places that aren’t filled in – in the striped areas, so few people filled out the survey that StatCan can’t release any detailed data about them due to concerns about data quality and confidentiality. Given how much policy development and research takes place with these numbers, these blank spots on the map have significant implications for the future of the communities that lie within them. We can only hope that the federal government seriously reconsiders this policy.