In early June, I wrote this post about the loss of the long-form census and what it means for those that produce and analyze census data to better understand demographics and plan for Canadians. We’re now finding just how destructive the decision to stop the long-form mandatory census was.

In my last post, I examined the Sun Media op-ed “Census cry was all elitist paranoia”, and how the author uses the roughest of population migration data to argue the data is unsullied, without offering any examination of the deeper issues the cancellation presented. Instead of examining skew and how the under- or over-representation different people groups makes the data utterly unreliable, he takes clumsy swipes at so-called “elites”, the only people that apparently care about the change.

Unfortunately this is only one example of insulting and uninformed opinion pieces on the issue. Another is this Sun editorial from February of this year, “Senseless census spluttering ceased”. In particular, this observation by the author:

Notice there are no screaming legions today trying to sell the notion that the census is now meaningless.

No, instead of carrying on with their condemnation of the mandatory long-form census’ departure, they are gleaning the important information the census continues to deliver.

As details of the census findings begin to roll out, there appears to be nothing of great significance lost for the marketers and the analysts to digest and, for governments, NGOs and the non-profits to formulate workable strategies for the future.

All that teeth gnashing was all for naught.

The author also argues that statisticians are quietly moving on with their business as usual with data that is as reliable as ever. Everything now being presented by Statistics Canada as they roll out the first set of data since the decision was made to scrap the traditional census completely contradicts such statements. Such editorials are now proving to be premature at best and at worst, purposefully misleading.

This weekend, these news reports came out, stating that the long-form census cancellation is rendering new StatsCan data unreliable and questionable. Halifax’s ChronicleHerald put it this way:

Data users are advised to exercise caution when evaluating trends related to mother tongue and home language that compare 2011 census data to those of previous censuses,” Statistics Canada states bluntly in a box included in its census material.

Those are strong words for a statistical agency, since they raise profound questions about how the data can be used reliably to come to conclusions about language trends. Officials have undertaken a thorough investigation, with a report to be published shortly.

As a student of information technology, I find the language StatsCan chose striking. It may not read as a glaring condemnation of the new methodology, but as the article observed, what has been said is very strong words for a statistics agency. For now, it seems that with some guessing (a word no statician should ever be found using!) and analysis of contextual evidence from past censuses, some approximation of reality can be gleaned. However, this in itself is a completely invalid method of producing data, and we will only be moving further into the wilderness of guesswork and supposition the more information comes out using this new voluntary method. As the Globe and Mail article stated,

…the method of gathering the new data makes it difficult to assess where Canada is going in comparison to where it has been. Experts say these questions will only grow more complicated as results from the voluntary survey start to roll in next year.

At the highest levels, this information will still be fairly reliable. As the Sun editorials observed, we can still see movement across the provinces and understand where growth is and isn’t happening. But this is far from a complete picture, and is far from the most valuable data. Professionals including urban planners, sociologists and community organizations rely on much finer, community-level data to do their jobs effectively, something that will become much more difficult as time goes on. Observations by Ivan Fellegi, the chief statistician of StatsCan until 2008 in this CBC article put it perfectly:

Even if big cities have high response rates, the swings in the data within smaller communities suggest to him that there may well be similar swings within certain groups in urban centres. Good data for the City of Toronto as a whole says little about how the Chinese population is faring, or whether low-income groups in the downtown core have enough daycare facilities.

My hope is that either the current federal government realizes just how damaging the decision to cancel the long-form census was for professionals working to understand the people they serve with the best possible data, or that the next government will reinstate the mandatory long-form census.

If you have an opinion on bringing back the mandatory long form census don’t forget to inform your MP, you can find their contact information here.