In fact, what would have the most success and be the best for us would be for the federal government to put systems in place to collect data about children, and for that data, which would be protected and high quality, to be hosted in data centres dispersed throughout Canada and be made available to all researchers.
Historically, we had access to very good data about children. It wasn't perfect, but it was very good quality. It came from the National Longitudinal Survey of Children and Youth, which was abandoned in 2008, at the time it was last published.
This type of data is extremely important for understanding children's development. We can't just rely on administrative data because, as other witnesses have said, administrative data does not show the child's cultural community or religious affiliation, for example. All sorts of dimensions are missing from administrative data, but they are very important for understanding children's development and welfare.
We need a central initiative that collects data from all over Canada so we can compare results from one province to another; that data set must be large enough to be able to study small groups. I heard other witnesses talk about the fact that we can't study small groups, but it's not because researchers don't want to. I assure you that we want to. However, when we are working with small samples, the data are not valid. When there are not enough people in the sample, we can't say anything about it and Statistics Canada doesn't even let us output the data.
I was a laboratory director at Statistics Canada for several years and I am very familiar with the machine. We need major funding from Statistics Canada in the area of children's health. Children represent almost 20% of our population, but I can assure you that Statistics Canada does not devote 20% of its budget to collecting data on children. The agency can put out the Labour Force Survey every month; it can...