It's a great question.
Creatively, we like to exploit existing datasets that others might have, which we can link to other datasets to answer some of those research questions. Ideally, as Mr. Clermont has noted, with regard to one of the questions about court cases and police cases and how they are being dealt with in the courts, what we've done periodically through a lot of work is link our police data with our court data so we can follow those cases, those victims and accused, through the justice system. The datasets aren't created for those linkages, but we do the job manually. What we would like to do perhaps is to look at offenders and where they are in the systems. Do we see them in hospital data? Are they filing taxes? They are likely not. With regard to victims, if they are domestic, can we find out who they are through ethnicities if NGOs have large datasets? Generally speaking, my understanding is that an NGO is focusing on services to their people as opposed to data collection. That being said, within Statistics Canada we are working with directors of victim services, provincial and territorial, and we often ask about human trafficking cases. Currently, they don't collect that data, but we often ask about it. It's a balancing act.