Skip to main content

Twice now, the Harper government has chosen not to support same-sex unions. Canadians expect betterArmando Franca/The Associated Press

An unexpectedly high number of same-sex marriages in places like the oilpatch left census-takers scratching their heads — until they realized many of the "couples" were only splitting the rent.

As a result, Statistics Canada said Wednesday, it may have overestimated by as many as 4,500 the number of same-sex married couples in parts of the country.

A last-minute discovery forced the agency to hold back some census data on gay and lesbian married couples when they realized they couldn't be certain if some people were hitched or only roommates.

So during Wednesday's release of the latest tranche of census data, Statistics Canada only provided same-sex marriage numbers for the country's larger cities and not for smaller communities, where the numbers might be off.

It all amounts to an asterisk in the latest batch of numbers about same-sex marriage in Canada.

"We observed that there was a possible over estimation of same-sex families," said census manager Marc Hamel. "The counts for some smaller communities seemed too high."

None of this changes the fact that more and more of Canada's gay and lesbian couples are tying the knot.

Statistics Canada said same-sex marriage nearly tripled between 2006 and 2011. The census data also shows the number of same-sex couples jumped 42.2 per cent during that same period.

The agency counted 21,015 married gay and lesbian couples and another 43,560 in common-law relationships. That's up from the 2006 census, which enumerated 45,345 of them — 7,465 married and 37,885 common-law.

One possible explanation for the error is people who leave home for transient work in places like Alberta and Saskatchewan, Mr. Hamel said.

Roommates who are married — just not to each other — could have been counted as a couple.

"It could be related to how some people enumerated themselves," Mr. Hamel said.

"We seem to observe that in more transient communities where we have a lot of temporary workers coming in. So it could be people living together, for example, and reporting each other as married, but not necessarily to each other."

Changes to the census form may also be partly to blame.

The 2006 census form — the first to count same-sex couples — didn't have a designated box for gays and lesbians to indicate they were married.

The 2011 census form did have a special check box for same-sex married couples. The differences between the two census forms may have had some bearing on the data, Mr. Hamel said.