Census, homelessness and gated communities

I am posting this on behalf of a colleague from Victoria B.C., Ian Faris, an employee with Statistics Canada for 20 years, and now a research analyst and member of the Canadian Social Data Consortium. The data consortium is organized to “liberate” census and related data at a modest fee for city planners, public health bodies, school boards, health districts, police and community-based agencies in municipalities across the country. The Canadian Council on Social Development has facilitated this process for the past two cycles of the Census, and keeps us connected with one another.

***************************************************

Good article in the Calgary Herald on Friday July 30th by Sarah McGinnis. But it contains one small piece of “misinformation”, unfortunately, from Susan McDaniel (University of Lethbridge Prentice Research Chair in Global Population and an adviser to Statistics Canada).

“The mandatory long form census is the only complete view of the population, allowing a rare opportunity to track the homeless, count those living in illegal suites or get details on the very rich living behind gated communities”

The Census does not actually do any of these things.

Statistics Canada attempts through all reasonable efforts to count the homeless at shelters and the like on Census Day. The agency acknowledges, though, that it is very difficult to count persons who do not have a permanent address or who consider themselves homeless and some people (for example, those who slept out of doors) may have been missed. Persons without a home and staying with family or friends at the time of the Census may have been included in the count for that household. Some limited data may be available for persons in “Shelters for persons lacking a fixed address”, but this will not be a complete count of the homeless by any means. The Census, which counts persons with a usual place of residence, thus, is not really the appropriate vehicle for collecting nor disseminating information on the country’s homeless. This is the case as of 2006. Maybe some day Census counts of the homeless may become more complete, or a new specially designed survey may be created for this purpose.

Basement suites in single-family dwellings are counted as “Apartment or flat in a duplex” and although a Planner may know that a certain percentage of such suites in a given area are illegal, they are not identified as such in the Census. Census data does not even allow users researching aggregate data to directly distinguish between the “suites” in the main floor of a house versus the “suites” in the basement and they have to estimate this for themselves using the Structural Type of Dwelling variable, the Tenure variable (i.e. Rented or Owned) and Household variables (such as number of persons in the household or living arrangements) in custom tabulations. The Census Reps do not have any way of identifying the legal status of a particular suite. They just count all suites. Maybe someday someone will devise a question, which will allow the Census to count secondary (let’s not call them “illegal” and needlessly worry homeowners) suites so we can get a truer measure of the available housing stock. Regulation of secondary suites is a municipal responsibility.

Similarly with gated communities, unless, such communities constitute an entire Dissemination Area or DA (expensive custom geographies notwithstanding) on their own, they cannot be identified separately from any other. And, again as stated above, of course all data for any persons living in these communities are only available in the aggregate and are subject to suppression rules. Using terms such as “gated communities” imparts a negative stereotype for some, provokes a defensive stance for others. The Census does not use such terminology and consistently reports information in a very neutral fashion.

I wish certain spokespeople would be a little more careful with how they represent what is collected by and available from the Census. Imprecise descriptions can attribute more to the Census then it is able to provide.

Incorrect and inappropriate use of certain terminology (as it applies to availability of Census data) actually also does a disservice to those making a case for the Census vis-à-vis confidentiality at enumeration time, etc. For example, oftentimes respondents are reluctant to tell the Census Rep that they have a suite in their basement because they fear that this information (revealing their “illegal” suite) will get back to their municipal government. This feeds the “intrusion” argument and contributes to an undercounting of these dwelling units. And so on.

And then this incorrect terminology will get erroneously applied to Census data and reported over and over again in the media. And over time it becomes “fact”.

The Census does not count the “homeless”, “illegal suites” or “gated communities”. The Census does not actually do any of these things. We may like it to, and wish that it did, provide us with this information, but it does not. Someday, perhaps, it will, but in the meantime let’s not ascribe to the Census that which it is not. Let’s just value it for what it is.
Ian Faris