The Commons: Verifiable data are no match for riddles and notions

A committee hearing on the long-form census spells out the epitaph for this whole sorry affair

It was Mike Lake, the Conservative backbencher doing his best to seem the humble servant simply conveying the wishes of the people, who seemed to give the game away. He was relating his consultations with the electors of Edmonton—Mill Woods—Beaumont. And here he noted that when some of his constituents were informed, by him, that not filling out the census could result in financial penalty or imprisonment they became, in his words, “quite agitated.”

It was Charlie Angus, the NDP member seated kitty corner to Mr. Lake, who pointed out what appeared a rather crucial sequence of events in this telling. It was here that the chain of causation seemed to become tangled.  And it was here that this whole sorry affair found an epitaph. For if we can say anything about the quinquennial census, perhaps it is this: not until it was made an issue, did it become an issue.

This morning of hearings, an industry minister and two former chief statisticians summoned to Parliament Hill to discuss the nature of data collection in a democratic society, was often so profound. And if the 2011 census is destined to be rendered useless to future generations, at least our descendants will have these two hours to tell them all they need to know about the state of this nation’s management as it embarks on the second decade of the 21st century.

It was the minister, Tony Clement, who went first, steadfastly defending a decision at least ten of his Twitter followers support. He lamented the intrusiveness, he bemoaned the coercion. He managed to conclude his opening statement with the following shell game of a sentence: “I encourage Canadians to fill out the national household survey should they choose to do so.” He seemed for the most part to be working on commission, paid per uses of the phrases “fair balance” and “jail.”

The latter was of particular concern: that Canadians might be imprisoned for failing to fully fill out the long-form census was quite obviously offensive to his side. That this government didn’t immediately, upon taking office in January 2006, call off that year’s census was left unaccounted for. That the Prime Minister has, so far, not apologized to Canadians for putting them through that apparently fascist exercise in state coercion and intrusiveness was left unexplained.

The more specific the question, the less Mr. Clement seemed interested in answering. Liberal Anthony Rota asked Mr. Clement how many Canadians had been imprisoned for failing to complete the census since that penalty was added to the Statistics Act in 1971. Thrice he pestered the minister with this question. Not once, at least to these ears, did the minister concede that not a single Canadian has ever been so punished.

This could all be traced back, Mr. Lake offered, to a Liberal government’s decision in 1971 to expand the census to include a long-form questionnaire. Except that, as one of the former chief statisticians later explained, the long-form has essentially always existed and that the change in 1971 had only to do with the introduction of a separate short form.

Mr. Clement said he was willing to listen, that he was open to other options. But he could not explain why he had not yet met with a group of civic, economic and academic leaders who had formally requested an opportunity to discuss the matter with him. (Nor did he address the National Statistics Council’s allegation that he would not meet with them to consider their proposal for compromise.)

The minister claimed principle. He leaned forward and pleaded. He sat back and rocked in his chair. He kept to his belief that his government had found a balance between the need for data and the apparent burden of supplying that data. He repeatedly asserted the government’s ownership of the decision to change the census process. “We take full accountability,” he said.

This only after repeatedly and explicitly implicating Statistics Canada and Munir Sheikh, the former chief statistician, in the change, thus violating a principle of governance that brought about the end of Mr. Sheikh’s 34-year career in the public service.

A half hour after Mr. Clement finished, the meeting reconvened to hear from Mr. Sheikh and his predecessor, Ivan Fellegi. And it was here that the yawning pit of emptiness in the middle of this discussion was most obviously displayed.

Mr. Sheikh, until recently a man dedicated to anonymous service, seemed entirely unsuited for this situation. Rather than explain his own resignation, he read someone else’s explanation, choking up as he did. Asked if it was Mr. Clement’s comments that had led to his exit, he said it was media reporting that brought about the end of his career, leaving unsaid that the story in question was explicitly based on Mr. Clement’s comments.

In a small, accented voice, Mr. Sheikh explained the important difference between voluntary and mandatory surveys—that the former does not provide the same quality of data gleaned from the latter. Undaunted, Conservative Tom Lukiwski attempted to persuade Mr. Fellegi that since Statistics Canada conducts some voluntary surveys, it should be fine with conducting a voluntary census. Mr. Fellegi calmly explained that voluntary surveys depend on the benchmarks provided by the mandatory census to correct miscalculations. Mr. Lukiwski continued regardless.

When it was his turn, the government side’s David Anderson opened with two questions, taken apparently from the long-form census, about home repair. Mr. Fellegi and Mr. Sheikh happily answered. If this had been some attempt to embarrass the two statisticians, Mr. Anderson had apparently overestimated his own wit.

Mr. Anderson proceeded to lament for the mandatory agricultural census, perhaps forgetting that his government has so far not done anything to change the threats of fine or imprisonment that accompany that survey. Whatever Mr. Clement was feeling this day about his government’s ownership of this decision, Mr. Anderson seemed quite intent on finding Statistics Canada’s onus. He attempted to press the statisticians on the specific content of the questions in question and the threat of imprisonment. And then he posed something of a riddle, wondering how the census could be considered mandatory if, as some have suggested, the threat of prison were eliminated.

The statisticians could not be roused to engage Mr. Anderson in a struggle. The questions, they said, depended on the demands of those who use the data (users, it was left unsaid, that include the federal government). As statisticians, they were here only to speak to the statistical process.

One last note was then drawn out by the Bloc’s Richard Nadeau: the questions on the census itself were included only after the approval of the government of the day. This seemed perhaps the point.

Indeed, if the source of this present debate is not great public outrage with the machinery of government, the explanation would seem to be simply this. That what we see here, and what we saw today, is simply a government made greatly uncomfortable by its own role and place in society.