I would just add that Statistics Canada has excellent statistics; it's a question of the way you use them, what you look for. There is no such thing as an innocent number.
If you're only looking at very simple disaggregations, you're going to get a certain evidence base. If you complicate the story and look at other factors, you're going to get another evidence base. Researchers across this country are showing the difference when you complicate the story.
So it's a question of relationship with researchers who know how to use these databases more than of just saying that it's either good enough or not good enough.