Methodology: Crime and violence in Latin America and the Caribbean

Latin America and the Caribbean is one of the regions most affected by violence and crime, but at the same time, has little available information about this problem. When data exist, it is not always presented in a comprehensive way, and usually suffers from reliability problems.

The report Crime and Violence in Latin America and the Caribbean is a translation of the publication written in Spanish Delito y Violencia en América Latina y el Caribe, where the reader can find a complete analysis, methodological notes, and a statistic Annex. Whereas this English version contains only the introduction, with a description of the main findings, and the “Country Profiles”, which provide detailed information about each country.

One of the information sources in this work is the United Nations Office on Drugs and Crime (UNODC), which primarily receives crime statistics from National Polices or Justice Ministries. Another source is the Pan American Health Organization (PAHO), which collects mortality data from each country, including violent deaths, provided by national health systems, with a detail of event characteristics, as well as the victims’ age and sex.

The last source was the Americas Barometer survey – Lapop, coordinated by Vanderbilt University in 2012, which contains questions on personal victimization and citizens’ opinions on security, justice and the police in 24 countries from the region. The survey contains a sample of 38.631 cases.

Regarding the three sources, we not only verified the internal consistency of their data, but we also compared them in order to identify problems or understand their differences. In the Lapop and PAHO cases, we also worked directly with their databases.

In addition, this paper conducts statistical analyses to look for tentative explanations of, among other things, victimization levels, fear of crime, trust in the community and the police. We control these variables among themselves as well as with socio-demographic factors at the individual and aggregate (country) levels of analysis, conducting a cross-sectional comparison of countries, which can be found in the full Spanish version.

It is noteworthy that, from the Caribbean region, we only take into account the cases of Haiti, Jamaica, Dominican Republic, and Trinidad and Tobago. Other countries or islands were not considered due to lack of sufficient information. In any case, as they tend to be small populations, their absence does not significantly change the aggregate analysis. Anyway, to compensate their exclusion from the body of this work, the Statistical Annex contains some information about homicides on those islands.

Get the data: How was the 99% increase in homicides in Central America calculated?

Did you find this story interesting? Please support AOAV's work and donate.
Donate