| Posted: July 26th, 2013
Over the last year, I have written several articles about racial disparities in findings that a homicide was justified. Recently, two pieces in particular have captured some attention, one published by PBS Frontline, and one on this blog. Many readers have expressed interest in learning more about the methods I used to arrive at the conclusion that the odds a black-on-white shooting is ruled justifiable is less than 10 percent of the odds that a white-on-black shooting is ruled to be justified.
Today, I have a full report that documents the methods used in previous posts and that includes all the study results. For the skeptics, you can download the data yourself here, and if you email the public affairs team, we will send you the SAS code I used to perform the analysis.
There are two items to note in the new analysis. First, the analysis includes an additional year of data (2010) that was not available at the time the PBS Frontline piece was written. That article reports my finding that the odds a white-on-black homicide was ruled to be justified were 254 percent higher than white-on-white homicides. The new analysis finds the odds are 281 percent higher. The difference, I believe, is due to a trend in justifiable homicide findings. A regression analysis of FBI Supplementary Homicide Reports from 2005 to 2010 finds that the odds a homicide is ruled to be justified have increased about 10 percent a year, which validates the change in the new data.
Finally, one bit of detail from the analysis of the FBI data that is important to understand the findings: In the FBI data, a homicide is said to be justified if it is coded as the “killing of a felon by a private citizen”; that is, in order for a law enforcement finding of self-defense, it must be the case that the perpetrator feared for his or her life. If you give anyone a reason to fear for their life, you are committing a felony.
Photo by Tim Meko, Urban InstituteCrime, People |Tags: homicide, race, stand your ground, Urban Institute
1 Comment »