children playing

When Harper killed the Census he robbed Canadians

Printer-friendly versionSend by emailPDF version
Haider, Murtaza
Publication Date: 
24 Jul 2014


An ad in the Globe and Mail reveals the extent of harm the Harper Conservatives have inflicted on Statistics Canada. [Read online at The Huffington Post to see a copy of the ad.] Because of poor quality, Statistics Canada is not releasing data at finer spatial scales because the Harper Conservatives killed the mandatory long-form Census and replaced it with a voluntary survey of dubious quality.

The Conservatives' attack on the Census not only cost Canadians their most valuable source of information, it also led to the departure of Canada's leading civil servant, Dr. Munir Sheikh, who in July 2010 resigned his position as Canada's chief statistician to protect the integrity of the institution he led.

It happened exactly four years ago with the government's decision to eliminate the long-form Census that collected information about housing, transport, employment, income and other key attributes required by various tiers of governments to plan and deliver public services. The Conservative government's decision has cost millions more to taxpayers to collect poor quality data. Even worse, now Canadians have to pay even more to private vendors to get the same or better data.

Academics, economists, statisticians and others unanimously criticized the government's flawed decision to eliminate the long-form Census. They were of the view that eliminating the long form would create a permanent break in the Census because the change in survey methodology would make it impossible to conduct any longitudinal study comparing statistics over time. Furthermore, the voluntary survey would result in sampling errors and self-selection biases that could not be fixed.

The government cited unsubstantiated privacy concerns to eliminate the long-form, which, however, did not make any sense. The government wanted to replace the long-form Census, which is administered to a 20 per cent random sample, with a voluntary survey to be administered to 33 per cent of Canadians. How asking the same questions to a much larger population would address privacy concerns defied logic and exposed the complete lack of reason in the thinking of Canada's Conservative government, which has since then continued to weaken Statistics Canada by cutting its funding.

Dr. Sheikh, being the head of Canada's statistical agency, advised the government that no matter how large the sample size of the voluntary survey may be, it could not replace a mandatory Census. His concerns were ignored. Dr. Sheikh ultimately resigned.

It is, however, important to note that Dr. Sheikh resigned not because the government failed to listen to its top bureaucrat. He respected the government's authority to ignore the advice of civil servants. Instead, he resigned because the government tried to misrepresent facts to Canadians about the accuracy of the planned voluntary survey.

Canada's then Industry Minster, Tony Clement, told the news media in July 2010 that Statistics Canada had assured him "[I]f you do these extra things: the extra advertising and the extra sample size, then yes, we [Statistics Canada] can do our job." Dr. Sheikh had never advised the government as such. His advice was exactly the opposite, which he repeated for the record in his resignation letter:

"Let me first of all say that it is the right of the government to make decisions, which if lawful should be implemented by any department of the government... The fact that in the media and in the public that there was this perception that Statistics Canada was supporting a decision that no statistician would, it really casts doubt on the integrity of that agency, and I as head of that agency cannot survive in that job."

Therefore, four years hence, we have the proof. The data are no good. It does not mean that NHS data are useless. It means that the data are inferior in quality than the one collected in earlier Census. Even more ridiculous is the fact that the poor quality data has cost Canadians millions more than it would have had the government not interfered with the Census. The Auditor General of Canada reported that the 2011 NHS cost Canadians $22 million more than the long-form it replaced and produced inferior data.

So not only did the taxpayers paid more for poor quality data, government departments and not-for-profits have to dole out even more money to private data vendors to gain access to the same data. A lose-lose situation for Canadians, which still has not dissuaded the Conservatives from repeating the same mistake in 2016.

At the fourth anniversary of the demise of Canada's long-form Census, Canadians should know that a lot is at stake at the next federal elections. The Conservatives' refusal to acknowledge their mistakes and correct the course has hurt the integrity of national institutions that Canadians cherish. Voters should remember this when they head to polls in the next federal elections.

- Read online at the Huffington Post