And yet when I read some of the testimony previously, and as I've appeared here over the years, the statistics are only as good as the information and the statisticians. So we're never going to solve this problem unless we're all on the same song sheet regarding the questions we want to find out.
For instance, take your fraud example, where it's a global fraud but there are ten victims. To reflect that there are many victims of one particular offence, you have to frame your statistical question in that regard.
So statisticians who are looking for information often--and I'm not being critical--don't know what the question is. Where are these statistics going to end up? Are they going to end up in the House of Commons to try to interpret a bigger picture, or are they in response to a narrow question?
So I think before we can rely on statistics, we have to almost give a mandate to the person we want to get the information to tell them what we want to know. I often feel that statistics are rolled out to suggest that crime is increasing when the evidence may very well be that crime is decreasing--some crimes. What crimes are we talking about?
I think we need to have specific information so that the statistics are reliable. I'm not sure--and I don't think you are satisfied--that the statistics you're getting give you the answer. That's because of what the question is and what the mandate is of the statistician. They may be looking for information on one thing and we're using it for another.