Macdonald-Laurier Institute Senior Fellow Philip Cross spoke to several media outlets on Friday about the error Statistics Canada made in calculating its July jobs figures.
The agency announced early last week that it had made a mistake in calculating its jobs figures for the previous month. On Friday it revised the number of jobs gained to 42,000, an increase over the 200 job increases it previously reported.
Some spent the week speculating that the error was made as a result of one of two factors: recent cutbacks in funding to the agency, or that the government’s decision to cut make the long-form census voluntary is now making it difficult for Stats Can to collect proper data.
Cross says both of these explanations are wrong.
Cross, who previously worked as Statistics Canada’s chief economic analyst, told the Toronto Star that the agency has all the resources it needs to conduct important studies like the Labour Force Survey.
Speaking with The Canadian Press, meanwhile, he said the mistake has nothing to do with the long-form census. That’s because the LFS relies on the short-form census, he says, which is still mandatory.
Instead, Cross says, this debacle should serve as a reminder that human error can still play a role in the collection and processing of data.
He told the National Post “this isn’t symptomatic of budget cuts, it’s not symptomatic of the census. It’s just somebody made a mistake”.
That’s why, he told the Metro chain of newspapers, Stats Can now needs to look at its system of checks and balances to ensure the mistake doesn’t happen again. This thought was echoed in a National Post op-ed on the subject, which cited Cross’s column.
On the whole, though, Cross says Stats Can is doing a good job at data collection and processing. As he wrote in an op-ed for the Globe and Mail last week, data quality at Statistics Canada is higher than it’s ever been.