By Brian Lee Crowley, National Post, February 28, 2011
Just how bad is crime in Canada... and how can we tell? Since our institute published Scott Newark's paper on Canadian crime statistics, many critics, including John Moore ("Having fun with Numbers", Feb. 23) have indignantly argued that there is a single objective reality about crime and it is adequately captured by Statistics Canada's annual Juristat report. But neither proposition is correct. And StatsCan agrees.
StatsCan acknowledges that there are many legitimate ways of reporting crime, and whichever you choose will affect the outcome. According to its handbook on justice statistics: "There are a number of ways of measuring the incidence of crime and each method will yield a different result."
So the choice is not between a "correct" method, and all others. Every method, including StatsCan's, affects the resulting picture about crime. StatsCan's analysts accept this, and they have not simply chosen to defend their practices. On the contrary, despite some disagreements, they have acknowledged that there is merit in a number of the questions that Newark has raised, and they have invited him to work with them to improve crime statistics reporting.
Of course StatsCan's imperfect way of measuring crime might still be better than the other options. Here are just some of the reasons Newark disagrees, and why the official picture of crime obscures as much as it reveals.
The 2009 StatsCan Juristat report, to pick just a few examples, tallies incidents of "causing a disturbance" but does not provide this information on much more important crimes like first-or second-degree murder. Similarly, one has to consult a much lower profile government publication to learn that in 2009, "There were 78 youth aged 12 to 17 accused of committing homicide in 2009, 23 more than the previous year. This represents the second-highest rate per 100,000 population reported in over 30 years." We think that StatsCan's flagship publication on crime in Canada needs to provide more accurate, thorough and relevant information to Canadians. Newark's paper helps show the way.
Moore and others have complained that the Newark paper focuses more on crime volumes than rates. Volume refers to the total number of crimes committed overall; crime rates looks at the number of crimes per 100,000 people.
But if these critics actually read his report, they would know that Newark recognizes the difference and agrees crime rates matter. By contrast, his critics ignore the fact that crime volumes also matter. If you walk every day on a downtown street in Toronto or Montreal or Vancouver, and you know that last year there were four gunshot deaths there and this year there have been six, you might not be terribly moved by officials telling you that there were also more people on the street this year, and so crime is obviously "down." There is now measurably more crime that worries you in your immediate surroundings.
It is perfectly fine to have a discussion about the relative merits of crime rates vs. volumes in forming a picture of crime; it is not fine to attack people for having the temerity to take the available data and ask whether we can do a better job measuring and analyzing crime.
The critics also claim that Newark's report supports an indiscriminate "tough on crime" agenda. They should actually read the report. The author repudiates such an approach, pleading for a better analysis and reporting on crime so that we can focus enforcement on the small number of people committing most of the crimes that really matter to Canadians. As he writes in the report: ". instead of being 'tough' on crime, it's better to be honest about crime so as to be smart about crime."
Finally, Newark's critics leave the impression that the Juristat report indisputably shows crime declining in Canada. But what does the report itself tell us about crimes that Canadians are likely to think are directly relevant to their own safety and that of their family? Among other things, that from 2008 to 2009:
- kidnapping/unlawful confinement increased by 76 incidents
- homicide and attempted murder increased by 84 incidents
-child pornography increased by 205 incidents
- using/pointing/discharging a firearm increased by 237 incidents
- trafficking of drugs other than cocaine and marijuana increased by 582 incidents
- sex assaults against children increased by 1,185 incidents
The vehemence of the reaction by the academic crime policy establishment to Newark's report is totally disproportionate to the substance of their criticism, which is mostly based on what they wish he had said rather than what he actually said. We will continue to work with researchers and StatsCan to create as accurate and informative a picture as possible of crime in Canada so that we can truly be smart about crime.
Brian Lee Crowley is the managing director of the Macdonald-Laurier Institute, an independent and non-partisan national public policy think tank in Ottawa. First appeared in the National Post.