Data for Brazilian police killings and victimization

Read full text Summarise with VIRAT
Authors: Kaizô Iwakami Beltrão
Cited by: 0

Summary

No summary yet. VIRAT can read this paper and write a plain-language summary of its aim, methods, key findings and limitations.

Summarise with VIRAT Free to try with your Science Catalogs account.

Abstract

This data support a study that examines lethal police violence and police officer victimization in five Brazilian states—Bahia, Espírito Santo, Minas Gerais, Rio de Janeiro, and São Paulo—from the 1990s to 2024. Drawing on mortality data from Brazil’s Ministry of Health (SIM) and police records compiled by the Brazilian Public Security Forum, the paper conceptualizes police killings as potential state crimes occurring within a democratic yet institutionally fragile context. Two dependent variables are analysed: police killings per 100,000 inhabitants and police officers’ homicide rates per 100,000 officers. Using OLS regression with backward elimination, the analysis tests socioeconomic, institutional, and structural covariates, including inequality (Gini), Human Development Index (HDI) dimensions, urbanization, GDP per capita, public security expenditures, police personnel, cargo theft, and youth male unemployment. Results show strong associations between police killings and inequality, HDI income and longevity, urbanization, and cargo theft. Police officer victimization rates are positively associated with police killings, suggesting a feedback dynamic in which violence reinforces further violence. The findings indicate that lethal policing is embedded in broader patterns of inequality, weak accountability, and contested state authority. The study contributes to debates on democratic governance, public security, and state violence by highlighting the reciprocal relationship between police lethality and officer victimization.

Fields of Study & Subjects

Policing Practices and Perceptions Crime Illicit Activities and Governance Criminal Justice and Penology

Publication details

Journal / source
Harvard Dataverse
Publication date
22 March 2026
Language
EN
Document type
Dataset