Sorry, but this census thing continues to intrigue me.

The census has changed since 1851 when it was first administered in pre-confederation Canada, but it has never been limited to simple questions about name, age, sex and place of residence. That early census asked about: occupation, religion, marital status, race (“coloured” or “Native”), education level, and infirmities (blind, deaf and dumb, or lunatic). It also asked many questions about physical dwelling space (construction materials, number of stories, number of families residing there, etc.). In 1861 “mulatto” was added to race options, “idiot” to infirmities, and literacy to education (persons over 20 who could not read and write). In 1871 the census asked no questions about race and asked only if one was “of unsound mind”. But it divided the literacy question into “cannot read”, “cannot write”.

In 1901 immigration and naturalization questions were introduced, and detailed employment questions were added. French and English were distinguished, and the category “from childhood” added to infirmaries. In 1911 the census was expanded again to include more employment details, and Infirmities were detailed as: blind, deaf and dumb, crazy or lunatic, idiotic or silly. And in 1916 mother tongue was added, reflecting the increased immigration from various parts of the world.

In 1971 major changes were made. The census was divided into a short form comprised of basic population questions plus nine housing questions, and a long form comprised of the short form questions plus an additional twenty housing question and thirty socio-economic questions. The former was sent to 2/3 of Canadian households; the latter to the remaining 1/3. Also, for the first time, in response to privacy concerns, people were allowed to fill out the questionnaire themselves rather than having to report their answers to a census worker.

In 1996 the short form was reduced in size and new questions were added to the long form including Aboriginal identity, population group, household activities or unpaid work, and mode of transportation to work. That year 80% of Canadian households were to complete the short questionnaire which now had only seven questions, while the remaining 20% completed the long questionnaire with fifty-five questions. 2001 saw questions about common-law relationships introduced, and in 2006 same-sex marriages and common-law relationships were added.

The census, as we might expect, has evolved over time. We no longer use the words “idiot”, “lunatic”, at least on official forms, we self report, and we sample 20% of households for most questions rather than requiring all households to answer all questions. These changes reflect shifting social attitudes and developments in the science of demographics. And every change has been accompanied by concern for the quality and continuity of the data. But changes have, and always will come.

Today the concern is about privacy (the right of individuals to refuse to answer questions they consider too personal or overly intrusive). – Not to be confused with confidentiality (the right of individuals to limit the sharing of their personal data once it has been collected). – And the privacy concern is not just about whether people should be allowed to withhold information, but the fact that an increasing number of people do withhold it. In the 2006 census 168,000 households refused to submit the form and many thousands submitted blatantly dubious data, claiming to be Jedi Knights for example. Indeed, one of the problems with compulsory surveys is that false answers are often submitted precisely because blanks are not allowed.

Personally I think it’s time for the politicians in Ottawa to give the whole matter a rethink. Do we need to force people to answer questions they do not want to answer? Do we need to simplify the process? Should we make the entire census, the long and the short of it, voluntary? In this day of constant data gathering, do we even need a census? StatsCan says we do, and the Brita folks say we need to filter the water, but I’m not so sure.