IIn 2010, the federal government decided to remove a whole series of statistical data from the legally required collection of census data in Canada. Since the first national census in 1871, Canadians have been asked to state their religion on the national census. This is an obviously useful tool for historians of religion as well as for a host of other scholars and others. In recent years the census was divided into two forms: a short form asking basic demographic questions every five years and a long form census sent using random sampling techniques to roughly 19% of the population. Random sampling - an effective technique used in a sophisticated fashion in the social sciences and Humanities - was abandoned in the 2011 census conducted in May of that year. While the National Households Survey, the questionnaire designed to replace the mandatory long form census, was sent to a variety of households, it is entirely voluntary. This renders it useless, albeit interesting, to scholars.
As the chief statistician of Canada Munir Sheikh, said briefly and to the point after his resignation in protest over the voluntary nature of the NHS in 2010:
“I want to take this opportunity to comment on a technical statistical issue which has become the subject of media discussion ... the question of whether a voluntary survey can become a substitute for a mandatory census ... It cannot.”