Behind the numbers: Taking Census: privacy versus the public good

The number for this story is 95, as in 95 percent. The figure represents the percentage of people who return the long-form questionnaire that is part of the census that up until now, one in every five Canadians was obliged to fill out. Statistics Canada distributed the questionnaire to probe for details the regular census form fails to capture. The long form delves more deeply, collecting information about what we believe, our ethnicity, how we move around. In short, how we live our lives. Because Canadians must fill out the long form (in addition to the short form, of course), most end up doing just that, meaning that 95 percent return the forms to StatsCan. The information is good as gold to researchers, who have preferred to do their jobs quietly. Until now!

But before discussing these new militants and why they're so upset, let's have a quick word about the forms that are causing all the fuss.

The agency uses a rigorous methodology to select individuals who answer the questions on the long form. The selection process ensures that an accurate cross section of citizens who provide richer detail about their backgrounds. Researchers can then use that information in combination with other data sets to explain certain behaviour. For instance, do the prescription drug habits of people in a certain neighborhood have anything to do with socio-economic factors such as how much money they earn, or their ethnic background? Combining information from the long form with the actual prescription drug habits provides researchers in British Columbia with possible answers.

So if the information is so valuable, why is the government doing away with the census long form and replacing it with a survey that will cull similar information, but voluntarily? The government says it is simply responding to concerns of individuals who were worried about the intrusive nature of the questions.

Citing privacy concerns, the government will still allow StatsCan to distribute long form, but as a survey people can fill out if they wish.

"Canadians, I believe, and we believe, will be very happy to fill in the long form," explained Conservative senator Marjory LeBreton.

Not so, retort researchers such as Raymond Currie, who knows a thing or two about the methodology behind census taking. Currie is stepping down as the executive director of the Canadian Research Data Centre Network, a coalition of 24 (and soon to be 26) research centres at universities across the country. The network uses information from the long form for much of its work. The irony here is that the network also receives government money to conduct that research. So, on the one hand, the Ottawa is giving it money to conduct research; and on the other taking away some of the very data it needs to do that work.

And what kind of work? Currie uses the following example.

"Let's say an Italian community in Montreal wants to set up a centre for the elderly. Where do the Italian elderly live in Montreal? How many are there? Is there sufficient need or demand for a centre for the elderly of Italian descent who would like to be surrounded by people of their own ethnicity? You couldn't get that information if you didn't have a long form, which tells you all about minority groups."

Currie's group has sent a letter to Tony Clement, the federal minister responsible for StatsCan, demanding that the government reverse its decision. Currie echoes the concerns of researchers who explain that if you make something voluntary, then many people just won't bother, no matter how noble the exercise, and no matter how much you plead with them to cooperate. So even if the government plans to send the forms to a greater number of households, Currie and many other researchers who have spoken out (including former statisticians with the agency) argue that the return rate will still be well below the gold standard of 95 per cent. He says this means the data would be unreliable because it would fail to represent an accurate cross section of society, in large part because research demonstrates that certain individuals such as those from low-income backgrounds and immigrant communities avoid filling out forms. And these are some of the very people who would benefit from research into certain behaviours.

Currie says just because people dislike completing forms, doesn't mean governments should let them off the hook.

"People objected to seatbelts. People object to legislation on anti-smoking. They object to the legislation on cell-phone use in cars. But we do things for the common good. We have to measure the common good against the individual good. Nobody says we should cut out jury duty and not have it obligatory. And no one says we should cut out the census except the minister."

So far those arguments don't seem to be swaying the government. Standing beside Marjory LeBreton at the news conference in Ottawa was John Baird, the minister who has become the de facto spokesperson for the prime minister when he's not around. When asked about the long form and the reason for changing the government's policy, he gave an answer that left journalists scratching their heads.

"(The) government threatening to put people in jail if they don't tell how many bathrooms they have is a bit heavy handed and a bit ridiculous. We will respect peoples' right to privacy. At the same time the important information that's needed in the census will be gathered."

Flummoxed, journalists pressed Baird and LeBreton to explain what they meant by throwing "people in jail." (Listen to the audio here.)

LeBreton: "...as John said, (we) will not be throwing people in jail and threatening them with fines and chasing them around forever for not filling out the long form."

Baird: "I just think a lot of Canadians find it really offensive that big government steps in their lives, asking how many bathrooms they have in their house. I think it's ridiculous."

(UPDATE: I've just heard from an expert who pointed out that the 2006 long-form census didn't actually ask about bathrooms.)

So what advice, did the government receive? Who's complaining about the government being intrusive? And how did the government arrive at this new policy? The journalists had no luck obtaining answers. Baird suggested they put their questions to Tony Clement. So I called. No response so far. However, in the past he has been quoted as saying that the government doesn't have to shout it from the roof tops every time it makes a change.

Undaunted, a number of groups are adding their voices to the disenchanted. Marianne Levitsky is one of them. Her field is occupational health and safety. She doesn't use the data from the long form, but felt compelled to express her concerns. She assumed there would be a petition, but after conducting a bit of research discovered this was not the case.

Levitsky considers herself just an ordinary Canadian who was so outraged that she also started a Facebook page and petition with 4811 signatures) which she is sending to Tony Clement.

She says Baird's comment about people being sent to jail is ridiculous and feels that many Canadians would gladly cooperate, as they have in the past.

"Most Canadians are glad to fulfill their obligations of citizenship. And it's not a matter of doing it because you think you might be put in jail. And we usually do it willingly and gladly when we see there is a public benefit."

One of the people to sign her petition is Steve Morgan, a health policy researcher at the University of British Columbia.

"It sounds as though it's just bean counters at some levels," he says. "But when I can show you I can do a study that demonstrates that children from a minority group living in neighborhoods that are predominantly white, do worse off than children where there are larger concentrations of their ethnic identity. That's not to say that the policy prescription is to racially segregate. But it does tell us something about the experiences of people, and maybe help us talk to doctors about how to make that extra effort to be culturally competent when you're talking to a person who is a recent immigrant in a neighborhood for which there is not a lot of immigration."

Steve Morgan, Marianne Levitsky and Raymond Currie are three voices that seem to be part of a growing chorus of critics which includes key members of the business community that use the data; the country's official languages commissioner who is investigating the government's decision; municipal politicians who depend on the data to figure out how to provide services to certain neighborhoods; and, of course, federal opposition politicians.

So, as we say in journalism, this is a story that seems to have legs, especially since it involves data that has been tantamount to gold to some of the country's leading researchers.

If you have any information to share about this story, please feel free to contact me at david_mckie@cbc.ca.