CDR’s nonprofit mission is to help the development of a data-driven human services system. One way we accomplish this end, is by working to educate communities on publicly available data, and how monitoring it can help them address community problems. Today I am going to talk about data relevant to the debates around policing of racial minorities, and also review specific examples from my home town of Allegheny County, Pennsylvania.
Trigger Warning: In regards to current concerns with racial justice, especially in regards to law enforcement and black & brown communities, the data to review can be very intense. We are about to talk about literal matters of life and death, and matters with deep systemic issues. The data can be alarming, but there is value in looking at it, and seeing how it can be helpful in moving our world to the more just vision that we all have in mind.
Data Sources: There are two very good public sources for Pittsburgh homicide data.
First, is the Pittsburgh Post-Gazette’s Homicide Dashboard, which gives you very up to date information on homicides, since it is populated by their journalistic coverage of every homicide in the city (they respectfully also list the full name of all homicide victims besides the rising number of deaths in the city as a year progresses).
I would like to highlight that July is clearly the peak of this line chart, followed closely by August. That is to say: The summer months of July and August have racked up the highest homicide counts since this tracking began 15 years ago.
A second dashboard of homicide information is maintained by the Allegheny County Department of Health. This data usually lags a few months behind, but that is because it is much more scrupulously vetted for information about the homicide before it is reported: such as its location, the actual time of death, the victim’s race, cause of death, and other demographic information.
Through review of this data we can learn many things, like that most of the homicides are due to firearms (~83%), that most of the victims are African-American Males (~69%) and between the ages of 15-34 … that most of the homicides happen on weekends…. You can drill down quite a bit, including mapping that shows you the neighborhoods with the highest needs.
Also through comparing these dashboards you can see that currently Allegheny County has more homicides at this time than we did this time last year (52 right now, compared to 46 last year at the same time). With the dangerous months of July and August still in front of us, and the current stressors in our society from Covid, to unemployment, and to social unrest … this could be a bad year for homicides.
What do we do with this:
This data provides a base-rate for the most extreme form of community violence that we are trying to address in our society. Just like in traditional health research, where hospitalization is a general measure for the success or failure of a health intervention, the homicide rate of an area tells us the background level of community safety. Bending this curve would be a meaningful change for the community, and would represent more than just homicides reducing, but would also mean improvements in the general livability and prosperity of an area.
And in regards to questions of racial justice in policing, this rate is acutely important. To a certain extent, in most places, the “frontline” workers dealing with homicide problems are police officers. This approach is rendering the current outcomes: Approximately 100 dead people a year, and a community at odds with the only resource currently allotted to the problem. When thinking about improving policing, the homicide rate should serve as an important benchmark of success: can we reduce violent crime without police intervention, at least in its current form? Will police reduction mean spread of community violence, or its reduction? How can police interact with other forms of intervention (community health workers, violence interceptors, community services) to bring down community violence and build a vital community?
These are all strong questions, and ones that have answers. Answers that can be seen in the data, as long as we choose to attend to it, collaborate in our interpretations, and act on it in concert.