Recidivism among ex-combatants of illegal armed groups in Colombia: An analysis of participants in the reintegration process

This thesis studies the extent to which demographic, illegal armed group (IAG) and demobilisation history variables influenced reoffending and crime severity in a sample of 16,834 ex-combatants from IAGs in Colombia, who were registered into the reintegration programme (RP) in 2006 and were monitored during a two-year ‘time at risk’ period. The purpose of the study is to identify the main predictors of recidivism and crime severity among ex-combatants who were taking part in this reintegration process (Participants in the reintegration process – PRPs). In order to understand the variables that could predict reoffending and crime severity, the thesis will focus on key individual factors. The selection of variables is informed by prior research and based on information available in a dataset from the Colombian Agency for Reintegration (Agencia Colombiana para la Reintegración = ACR). The dependent variable is reoffending, and the selected predictor variables are ‘gender’, ‘age at the time of registration in the programme’, ‘ethnicity’, ‘ex-group of membership’ and ‘demobilisation type’. Furthermore, crime severity is analysed as an additional outcome. First, a comprehensive literature review was conducted and led to four additional contributions to knowledge. The main challenge for disarmament, demobilisation and reintegration (DDR) processes in peace-building societies is violence renewal and ex-combatants’ recidivism. More specifically, some components of the transitional justice framework could be a risk for the peace-building process and contribute to the recidivism of ex-combatants. Only a limited number of studies in Colombia and internationally have analysed the causes of recidivism among ex-combatants of IAGs, based on quantitative methods and empirical evidence (Chapter 3). Due to the gaps between criminology and peace and conflict studies this thesis creates a tailored framework to answer the research question of this thesis. The reoffending rate in the sample was 7.3 percent. Second, a binary logistic regression model was developed to identify the predictors associated with reoffending during the ‘time at risk’ incorporating all of the variables. The results suggest that the selected predictors make a significant contribution to the prediction of reoffending (x2 = 434.1, df = 15, p = 0.000). Third, a polytomous logistic regression model was designed to investigate further distinctions by outcome severity (x2 = 484.6, df = 30, p = 0.000) with the same predictors. In summary, the research has four comparative strengths: the longitudinal quantitative data available; a range of individual demographic, IAG and demobilisation history predictors of recidivism; a specific focus on the criminological methods to explain the most relevant causes of recidivism among ex-combatants; and a theoretical framework that links criminology, political science (i.e. peace and conflict) and law (i.e. transitional justice) to explain the criminal careers of ex-combatants. Important limitations were identified, namely difficulty to access the data; incomplete data; process-produced data; and selection of dependent variables. The variables that present an opportunity to make original contributions to the peace and conflict literature are the operationalisation of reoffending by including ‘time at risk’, crime severity linked to crime type after demobilisation, ethnicity, categories of ex-group membership, and demobilisation type.

Name dc:type.qualificationname

Doctor of Philosophy (PhD)

Level dc:type.qualificationlevel

Doctoral

Grantor dc:publisher.institution

Law faculty. Institute of Criminology

Year dc:date.issued

2020

Author dc:creator

• Triana Enemocon, Sergio

Advisor dc:contributor.advisor

• Eisner, Manuel

• Violence prevention
• Reintegration
• Recividivism
• Reoffending
• Risk assessment instrument
• Disarmament, Demobilisation and Reintegration
• Prediction
• Ex-combatants
• Non-State Armed Groups
• Cambridge Crime Harm Index
• Crime severity

Licence

• https://apollo8-f-pro.lib.cam.ac.uk/bitstreams/5200efde-1f18-4d05-8c77-a8062373184b/download
• https://www.rioxx.net/licenses/all-rights-reserved/

Language dc:language

eng

Author Identifier

0000-0001-5436-9282

OAI identifier oai:identifier

oai:www.repository.cam.ac.uk:1810/323415