In 2002, Colombia had the highest homicide rate of any Latin American country(Berkman, 2007). The origins of this violence, however, are complex and difficult to identify. It would be sensible to argue that it cannot be explained by any one particular factor, but rather an assortment of many factors that wholly represent the social, economic, and political conditions of Colombia. By better understanding the origins of Colombian violence, policy makers can more effectively address and alleviate this prolonged issue. This study examines the geographic nature of municipal homicide rates for Colombia in 2005. The purpose of this study is to determine whether there are any discernible patterns in the geographic distribution of homicide rates across Colombia at the municipal level. It also aims to determine what combination of statistically significant predictors, if any, generates acceptable regression models for predicting the distribution of homicide rates. Spatial autocorrelation methods, particularly Global and Local Moran’s I statistics, were used to identify the clusters of high-value homicide rates. Regression models, specifically OLS and GWR, were utilized to examine the relationships between homicide rates and an assortment of geographic factors, including Coca Cultivation Density, Presidential Election Participation Rate, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, andPublic Force Armed Actions Rate.
The results of this study indicate that clusters of high-value homicide rates were indeed located in the northern, southern, western, and central regions of Colombia. Among the aforementioned geographic factors, Coca Cultivation Density, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, and Public Force Armed Actions Rate all exhibited positive correlations. The variable exhibiting a negative correlation was the Presidential Election Participation Rate.
","abstract_html":"<p>In 2002, Colombia had the highest homicide rate of any Latin American country(Berkman, 2007). The origins of this violence, however, are complex and difficult to identify. It would be sensible to argue that it cannot be explained by any one particular factor, but rather an assortment of many factors that wholly represent the social, economic, and political conditions of Colombia. By better understanding the origins of Colombian violence, policy makers can more effectively address and alleviate this prolonged issue. This study examines the geographic nature of municipal homicide rates for Colombia in 2005. The purpose of this study is to determine whether there are any discernible patterns in the geographic distribution of homicide rates across Colombia at the municipal level. It also aims to determine what combination of statistically significant predictors, if any, generates acceptable regression models for predicting the distribution of homicide rates. Spatial autocorrelation methods, particularly Global and Local Moran’s I statistics, were used to identify the clusters of high-value homicide rates. Regression models, specifically OLS and GWR, were utilized to examine the relationships between homicide rates and an assortment of geographic factors, including Coca Cultivation Density<em>, <em>Presidential Election Participation Rate</em>, <em>Displaced Persons </em></em>Rate<em>, <em>Standard of Living Index</em>, <em>Terrain Ruggedness Index</em>, <em>FARC Armed Actions Rate</em>, </em>and<em>Public Force Armed Actions Rate</em>.</p> <p>The results of this study indicate that clusters of high-value homicide rates were indeed located in the northern, southern, western, and central regions of Colombia. Among the aforementioned geographic factors, <em>Coca Cultivation Density</em>, <em>Displaced </em>Persons Rate<em>, <em>Standard of Living Index</em>, <em>Terrain Ruggedness Index</em>, <em>FARC Armed Actions </em></em>Rate<em>, and <em>Public Force Armed Actions Rate </em>all exhibited positive correlations. The </em>variable exhibiting a negative correlation was the <em>Presidential Election Participation </em>Rate<em>.</em></p>","abstract_has_math":false,"creators":["Fowler, Brandon"],"institution":null,"degree_name":"Master of Science","degree_level":null,"degree_discipline":"Department of Geography and Geology","degree_department":null,"school":null,"contributors":["Dr. Jun Yan, Director, Dr. Margaret Gripshover, Dr. David Keeling, Mr. Kevin Cary"],"advisors":[],"committee_chairs":[],"committee_members":[],"year":2013,"date_issued":"2013-05-01T07:00:00Z","date_published":"2013-05-01T07:00:00Z","updated_at":"2026-07-24T06:08:16Z","subjects":["Geographic Information System (GIS)","Colombia","Violence","Political violence","Homicide","Social problems","Geographic Information Sciences","Latin American Studies","Other Social and Behavioral Sciences"],"languages":[],"rights":[],"rights_urls":[],"identifier_entries":[]},"links":{"outbound_url":"https://digitalcommons.wku.edu/theses/1229","outbound_label":"Repository record","outbound_source":"dc:identifier"},"metadata_groups":[{"id":"people","label":"People","entries":[{"key":"dc:contributor","label":"Contributor","values":["Dr. Jun Yan, Director, Dr. Margaret Gripshover, Dr. David Keeling, Mr. Kevin Cary"]},{"key":"dc:creator","label":"Author","values":["Fowler, Brandon"]}]},{"id":"academic_context","label":"Academic Context","entries":[{"key":"dc:type","label":"Dc Type","values":["Thesis"]},{"key":"thesis:degree_discipline","label":"Discipline","values":["Department of Geography and Geology"]},{"key":"thesis:degree_name","label":"Degree Name","values":["Master of Science"]}]},{"id":"subjects_keywords","label":"Subjects and Keywords","entries":[{"key":"dc:subject","label":"Dc Subject","values":["Geographic Information System (GIS)","Colombia","Violence","Political violence","Homicide","Social problems","Geographic Information Sciences","Latin American Studies","Other Social and Behavioral Sciences"]}]},{"id":"identifiers","label":"Identifiers","entries":[{"key":"dc:identifier","label":"Identifier","values":["https://digitalcommons.wku.edu/theses/1229"]}]},{"id":"additional","label":"Additional Metadata","entries":[{"key":"dc:description.abstract","label":"Abstract","values":["In 2002, Colombia had the highest homicide rate of any Latin American country(Berkman, 2007). The origins of this violence, however, are complex and difficult to identify. It would be sensible to argue that it cannot be explained by any one particular factor, but rather an assortment of many factors that wholly represent the social, economic, and political conditions of Colombia. By better understanding the origins of Colombian violence, policy makers can more effectively address and alleviate this prolonged issue. This study examines the geographic nature of municipal homicide rates for Colombia in 2005. The purpose of this study is to determine whether there are any discernible patterns in the geographic distribution of homicide rates across Colombia at the municipal level. It also aims to determine what combination of statistically significant predictors, if any, generates acceptable regression models for predicting the distribution of homicide rates. Spatial autocorrelation methods, particularly Global and Local Moran’s I statistics, were used to identify the clusters of high-value homicide rates. Regression models, specifically OLS and GWR, were utilized to examine the relationships between homicide rates and an assortment of geographic factors, including Coca Cultivation Density, Presidential Election Participation Rate, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, andPublic Force Armed Actions Rate.
The results of this study indicate that clusters of high-value homicide rates were indeed located in the northern, southern, western, and central regions of Colombia. Among the aforementioned geographic factors, Coca Cultivation Density, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, and Public Force Armed Actions Rate all exhibited positive correlations. The variable exhibiting a negative correlation was the Presidential Election Participation Rate.
"]},{"key":"dc:title","label":"Title","values":["Understanding Colombian Violence Through Geographic Information Systems and Statistical Approaches"]}]}],"canonical_facts":{"dc:contributor":["Dr. Jun Yan, Director, Dr. Margaret Gripshover, Dr. David Keeling, Mr. Kevin Cary"],"dc:creator":["Fowler, Brandon"],"dc:description.abstract":["In 2002, Colombia had the highest homicide rate of any Latin American country(Berkman, 2007). The origins of this violence, however, are complex and difficult to identify. It would be sensible to argue that it cannot be explained by any one particular factor, but rather an assortment of many factors that wholly represent the social, economic, and political conditions of Colombia. By better understanding the origins of Colombian violence, policy makers can more effectively address and alleviate this prolonged issue. This study examines the geographic nature of municipal homicide rates for Colombia in 2005. The purpose of this study is to determine whether there are any discernible patterns in the geographic distribution of homicide rates across Colombia at the municipal level. It also aims to determine what combination of statistically significant predictors, if any, generates acceptable regression models for predicting the distribution of homicide rates. Spatial autocorrelation methods, particularly Global and Local Moran’s I statistics, were used to identify the clusters of high-value homicide rates. Regression models, specifically OLS and GWR, were utilized to examine the relationships between homicide rates and an assortment of geographic factors, including Coca Cultivation Density, Presidential Election Participation Rate, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, andPublic Force Armed Actions Rate.
The results of this study indicate that clusters of high-value homicide rates were indeed located in the northern, southern, western, and central regions of Colombia. Among the aforementioned geographic factors, Coca Cultivation Density, Displaced Persons Rate, Standard of Living Index, Terrain Ruggedness Index, FARC Armed Actions Rate, and Public Force Armed Actions Rate all exhibited positive correlations. The variable exhibiting a negative correlation was the Presidential Election Participation Rate.
"],"dc:identifier":["https://digitalcommons.wku.edu/theses/1229"],"dc:subject":["Geographic Information System (GIS)","Colombia","Violence","Political violence","Homicide","Social problems","Geographic Information Sciences","Latin American Studies","Other Social and Behavioral Sciences"],"dc:title":["Understanding Colombian Violence Through Geographic Information Systems and Statistical Approaches"],"dc:type":["Thesis"],"thesis:degree_discipline":["Department of Geography and Geology"],"thesis:degree_name":["Master of Science"]},"updated_at":"2026-07-24T06:08:16Z"}