National Household Survey – the third kind of lie

In any attempt to seriously convince a person of something, it is vital that the proposal be backed up with numbers. Any claim can’t be backed up by math, expressed in a simple fraction or per cent simply won’t be as convincing. Whether it be in political races, abstinence campaigns or toothpaste advertising, statistics are used at every opportunity to sway the opinions of the uninformed.

oct.8.oped.censusMark Twain wrote that “There are three kinds of lies: lies, damned lies and statistics”. This remains true, especially when you consider the veritable erosion of Statistics Canada’s reputability.

Prior to 2010, Statistics Canada would conduct a nationwide census, which detailed various different types of info in both a long and short form questionnaire. These forms were mandatory, and result in fines or jail time for Canadians who failed to complete them. This was the primary way for Statistics Canada to collect detailed info on the Canadian populous which would then be used to support new legislation, programs, initiatives, etc.

At the same time, since the economic collapse of 2008, new interest was shown in the economic standings of Canadians, specifically the idea of income inequality. While some see this plight as merely one out of charity and “fairness”, it understood to be directly correlated to higher levels of child poverty, poor health and shorter lifespans.

Furthermore, income inequality negatively affects advocacy and engagement. As the middle class shrinks, those falling beneath have less resources to devote to democratic debate or community involvement. People shift from being contributing members to avoided burdens. If you’re still not convinced, new thinking supported by The Economist and the International Monetary Fund proposes that greater income equality may be necessary for long term economic growth.

Ideally, Statistics Canada would use standard, tried and true censuses to determine the true extent of income inequality in Canada. Unfortunately, in 2010 Prime Minister Stephen Harper cancelled the long form census. As soon as 2011, Stats Canada began releasing disclaimers with their reports, stating “data users are advised to exercise caution when evaluating trends […] that compare 2011 census data to those of previous censuses”.

Not only did the accuracy of their data begin to fall, but a voluntary replacement survey took the place of the mandatory long form census.

Named the “National Household Survey” (NHS), this voluntary survey has actually found that income equality is not as prevalent within Canada as one might have thought. For example, in the Toronto and Calgary metro areas, the inequality (stated as a value itself, known as a “Gini” number) has lowered by 12 per cent, going from .22 in the 2006 census to .19 in the 2011 NHS.

This would be great news, if the NHS wasn’t flawed by its very nature.

Unfortunately, experts tell us that voluntary surveys do not accurately represent a given populous, as one demographic is far more likely to answer it than another (see above: income inequality negatively affects democratic engagement).

Even the former head of Stats Canada, Munir Sheikh lacked any faith in the survey. As to whether it can replace a mandatory long form census, Munir  has said clearly “It can not”.

The truly worrying part about this is that these figures gained by the NHS will be used to influence others on behalf of political campaigns, government programming and anything else that it would apply to. But at this point, how could we possibly trust it?

This is yet another case of Canadian government, whether provincial or federal, failing to take proper and precise note of what is actually going on within our borders. See last week’s “Thrown Out Ontario court cases: can we at least keep track?” for similar negligence.

 

Pin It

Leave a Reply

Your email address will not be published. Required fields are marked *

* Copy This Password *

* Type Or Paste Password Here *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <strike> <strong>