The Relationship between Violent Political Rhetoric and Police Shootings

Background and Purpose: Police shootings of subjects/suspects/perpetrators has become a serious problem. According to the Gun Violence Archive, in 2020, a total of 2,226 persons were killed or injured, and in 2021, 2,222, persons were killed or injured by police in the U.S. Research on factors associated with such police shootings is an important area of research. Recently studies have shown a positive relationship between violent political rhetoric (VPR) and mass shootings in the U.S. This study investigated the possible relationship between VPR and police shootings of subjects/suspects/perpetrators in the U.S.

Methods: This study was a secondary analysis of police shooting data from the Fatal Encounters (FE) and Mapping Police Violence (MPV) data bases, and VPR data from an online book draft on violent political rhetoric. Two recorders obtained monthly counts of police shootings from the FE and MVP data bases, and data on VPR were abstracted from a graph in the online book draft, for the months from January 2014 through September 2019, a 69-month period. Researchers have suggested the numbers of police shootings may be undercounted and data from any single source may be incomplete, so the police shooting data from the FE and MPV data bases were averaged across counts. The data were analyzed using an auto-regressive distributive lag error correction (ARDL-EC) time series approach. The dependent variable was monthly rate of police shootings, and the independent variables were the monthly rate of VPR, a count of the monthly numbers of new gun applications, a variable used as a proxy for the numbers of guns owned, and changes in gross domestic product (ΔGDP), a variable found to be related to mass shootings in the U.S. in recent research.

Results: The results, based on the Pesaran, Shin, and Smith bounds test for a co-integrated relationship between VPR and police shootings were F = 15.4, p < .001, and t = -7.66, p < .001, outcomes indicative of a co-integrated relationship between VPR and monthly rates of police shootings, with R² = .50. The relationship between VPR and rate of police shootings was, b = .006, t = 2.24, p < .05; between gun ownership and police shootings, b = .10, t = .65, p > .05; and between VPR and ΔGDP, b = .08, t = 1.58, p > .05. The error correction term was 0.95, t = -7.66, p < .001, results indicating monthly deviations from any short term relationship between VPR and police shootings returned to the long run relationship within each month.

Conclusions and Implications: The results suggested that as VPR increases, so do the rates of police shootings of subjects/suspects/perpetrators. The results suggest social workers need to develop strategies, programs, and interventions to reduce VPR. The development and implementation of programs to humanize persons pursued by police, especially ethnically diverse and disadvantaged persons, are considered and discussed. This could be done in collaboration with various media outlets. Other possible programs and strategies are considered.