Working Paper: NBER ID: w24238
Authors: Roland G. Fryer Jr.
Abstract: Police use of force – particularly lethal force – is one of the most divisive issues of the twenty-first century. To understand the nexus of race, criminal justice, and police brutality, academics and journalists have begun to amass impressive datasets on Officer-Involved-Shootings (OIS). I compare the data and methods of three investigative journalism articles and two publications in the social sciences on a set of five rubrics and conclude that the stark differences between their findings are due to differences in what qualifies for a valid research design and not underlying differences in the datasets.
Keywords: No keywords provided
JEL Codes: J0; K0
Edges that are evidenced by causal inference methods are in orange, and the rest are in light blue.
| Cause | Effect |
|---|---|
| Racial differences in police shootings (J15) | Methodological differences (C90) |
| Race (J15) | Likelihood of being shot by police (J45) |
| Encounter characteristics and demographics (R20) | Likelihood of being shot by police (J45) |
| Suspect behavior (K42) | Likelihood of being shot by police (J45) |
| Race (J15) | Officer-involved shootings (OIS) (K42) |