I think this picks up on the other committee member's comment. Monsieur Clermont already commented a bit on what we do to validate the police-reported data. In terms of training police, we have online training for them to be able to understand how to correctly score the data, etc., and as we see from these numbers, a lot of the information is based on the willingness of people to come forward and report to police. We have seen in the past that when there are community outreach programs, numbers tend to go up.
I think in terms of working with police, continuing to sensitize them to the importance of the data, and training them, we find that it makes a big difference when a police service has a hate crime unit and has strong relationships with communities in terms of the willingness of victims to come forward. That's where we'll get information on the numbers.
The other aspect in which maybe there is a gap is that we know a lot about victims from the victimization survey, but we don't know a lot about offenders and what motivates them to commit these crimes. There is a gap in information in that respect, of understanding the thinking behind offenders' behaviour and why they may commit the crimes that they commit.