Crime statistics can be very useful in creating policy if one really understands them and does the work to dig past the basic numbers. Really, it is the same with all statistics. However, a long trend in day-to-day reporting in journalism is to present statistics as if they were a static, stand alone set of numbers that tell us what is going down. Even when a journalist takes the time to provide some explanation and context, headline writers simplify the main stats in a single line which is often misleading. Politicians then take that single line and run with it. They form campaigns around a single stat and push pet policy.
Nowhere is this trend more dangerous than it is with crime stats. One single number presented out of context can bring down a decent office holders career and help pass destructive policy. This is why it is important to dive deep into crime stats.
Last month, California released its homicide report for the year. It showed a small jump in homicides across the state. It made the news as a murder spike. Thing is that California is a very big and complex state. We have counties that have tens of millions of people and ones with fewer than 5,000. The mountain/foothill east is different from the Central Valley which is different than the coastal west. North and south are different, same with urban a rural. It is crazy to draw any conclusions about the state based on statewide stats that are provided with no explanation or context, especially when the stats do not reflect what is happening in every county.
Similarly, any conclusion based on the comparison of a stat from a single year with one of the following year is going to be bunk. We can see a jump in the number of murders one year and a steep decline in the next, but to understand what is happening, one has to step back and see what the trend is over the last 5, 10, 20 years. When you do that, you find that a one year increase in anything could very well be an anomaly, something that we won’t know for sure until the following year or better yet after five years have passed.
When we take that long view approach and go deep into the numbers for explanations then we can move beyond headline-drive politics and create policy that really works. We spare ourselves electing tough-on-crime reactionaries and lock-em-up policies, things that tend to make things worse than better. We also are able to see what is working and build on our successes.
Capitol Weekly just ran an analysis I did on California’s 2020 homicide rate. It goes deep but it is very readable. You will learn a bit about what to look for in stats and arm yourself with tools to approach stats like these in the future.