Data-driven insights into the Colombian conflict: a statistical analysis of child soldiers trends in the department of Antioquia, Colombia

Witnessing armed conflict and violence at an early age can become a source of trauma for young people, which is why the protection of children in armed conflict is a NATO priority.

Alice Farmer, a children’s rights researcher at Human Rights Watch, labelled child soldiering as “the Hunger Games of the Real World”. The reader of Suzanne Collin’s best-selling trilogy can recognise the issue at hand: many (although not all) of these children have been forced into the battlefield against their will and have to go to extremes to ensure their survival, often forced to choose between an opponent’s life and their own.

Troubling as this might be from a moral point of view, it is in the eyes of the legal system where the position of demobilised fighters is most ambiguous. The Minimum Age of Criminal Responsibility (MACR) in Colombia is 18, which prevents child soldiers from facing criminal charges and, simultaneously, falls short in fulfilling the rights of victims to justice.

Are these children victims who should be offered compensation and reinsertion into society, or are they perpetrators of serious crimes who ought to be held accountable for their actions?

This question has recently become of paramount relevance in Colombia following the signing of the Peace Accord in 2016 between the Colombian Government and the Revolutionary Armed Forces of Colombia (FARC). With its implementation currently underway, over 10,000 ex-combatants who were recruited as children are expected to demobilise and search for ways to re-enter society.

How can interdisciplinary data analytics help?

In collaboration with the University of La Sabana and supported by the Q-Step Centre at the University of Manchester, Trilateral Research is carrying out interdisciplinary data analytics with a view to identifying trends in the recruitment of child soldiers in Colombia.

The first step of this study consisted of translating, cleaning, harmonising and appending two datasets in order to build a unified data source.  One dataset holds data gathered by the Public Prosecutor’s office dealing with Transitional Justice; the other contains data collected by the Colombian ordinary justice system.

Both datasets provide a great deal of information on 660 illegally recruited children from the Colombian department of Antioquia (one of 32 sub-national administrative units Colombia is divided by), including their age, gender, educational level, municipality of residence as well as their motivation to join the armed group, the time spent as child soldiers and the reason for demobilising, among others.

Subsequently, the study will explore the relationship between municipality-level variables, (e.g., presence of illicit crops, multi-dimensional poverty, political violence, and land concentration), and individual-level variables, particularly the children’s motivation to join the armed groups. These motivations were classed as economic reasons, and family or political violence ones. To this end, machine learning algorithms that will take into account, among other factors, the effect of children’s municipality of residence on their propensity to join the armed group will be used.

The table below shows preliminary descriptive statistics of the sample of recruited children, following the data cleaning where the cases with missing data were removed. Children are mainly male (88%), have an average age of 15 years, and tend to have joined paramilitary groups (93%). The children who declared to have joined the armed group for family or political violence reasons are 35%, while those who claimed to have done so for economic reasons are 65%.

 

This research may inform a wider project (currently under evaluation) that aims to analyse illegal recruitment trends at the national level, considering all 32 Colombian departments. Furthermore, this study will appear in a publication on child soldier trends in Antioquia that will be included in a book, published by Palgrave Macmillan and edited by Trilateral Research. The book, inspired by Project Solebay, will focus on Modern Slavery and Human Trafficking in conflict.

 

Authors: Jose Vinaixa Kinnear, Q-Step Data Analyst Intern at Trilateral Research and Giuseppe Maio, Data Analyst at Trilateral Research

For more information about our work in this research area, contact our team: striad@trilateralresearch.com

 



‘Risk Assessment Report and Methodology’

You can view the Executive Summary and Table of contents of the Project Solebay Risk Assessment Methodology Report.

Please sign up to the Solebay mailing list to download the Full Solebay project report.