Community correlates of change: A mixed-effects assessment of shooting dynamics during COVID-19.

Nicole J Johnson,Caterina G Roman,Wenjia Zhang

doi:10.1371/journal.pone.0263777

Abstract

This study examines changes in gun violence at the census tract level in Philadelphia, PA before and after the onset of the COVID-19 pandemic. Piecewise generalized linear mixed effects models are used to test the relative impacts of social-structural and demographic factors, police activity, the presence of and proximity to drug markets, and physical incivilities on shooting changes between 2017 and June, 2021. Model results revealed that neighborhood structural characteristics like concentrated disadvantage and racial makeup, as well as proximity to drug markets and police activity were associated with higher shooting rates. Neighborhood drug market activity and police activity significantly predicted changes in shooting rates over time after the onset of COVID-19. This work demonstrates the importance of understanding whether there are unique factors that impact the susceptibility to exogenous shocks like the COVID-19 pandemic. The increasing risk of being in a neighborhood with an active drug market during the pandemic suggests efforts related to disrupting drug organizations, or otherwise curbing violence stemming from drug markets, may go a long way towards quelling citywide increases in gun violence.

Highlights

Declared a global pandemic by the World Health Organization in March 2020 [1], COVID-19 has had wide-reaching effects on many aspects of daily life for people around the globe
Using Seasonal Autoregressive Integrated Moving Average (SARIMA) models of several crime types in 16 U.S cities between 2016 and January 20, 2020, Ashby [5] found there was no evidence of significant changes in serious assaults based on comparison of the expected and observed frequency of assaults
This study explored the relative impact of several community features on rates of gun violence across neighborhoods and over time in Philadelphia during the COVID-19 pandemic

Summary

Introduction

Declared a global pandemic by the World Health Organization in March 2020 [1], COVID-19 has had wide-reaching effects on many aspects of daily life for people around the globe. Since the early days of the pandemic, criminologists have sought to describe and explain the patterning of crime in the wake of COVID-19, as a response to mitigation strategies such as stay-at-home orders [4,5,6] These efforts have been fruitful in discovering broad city-level trends that persist across space, and in identifying those that do not. Using Seasonal Autoregressive Integrated Moving Average (SARIMA) models of several crime types in 16 U.S cities between 2016 and January 20, 2020, Ashby [5] found there was no evidence of significant changes in serious assaults (either street assaults or those occurring in the home) based on comparison of the expected and observed frequency of assaults. Abrams [6] posited that homicides and shootings would likely follow different trends post-COVID-19, noting the contribution of gangs and drug-markets to violent crime rates

Methods

Results

Discussion

Conclusion