Date of Degree


Document Type


Degree Name



Criminal Justice


Joshua Freilich

Committee Members

Desmond Arias

Valerie West

Michael D. White

Subject Categories



Research in the social disorganization tradition has found community disadvantage to be one of the strongest and most consistent macro-level predictors of homicides in urban areas in the United States (Pratt & Cullen 2005). This dissertation empirically tests the applicability of ecological theories of crime to the spatial distribution of homicides in Bogota, Colombia, while proposing alternative measures of social disorganization that are analogous to those used in the American literature but that are more reflective of both social realities and data availability in Colombia. The study used data from several sources including official homicide figures from the National Institute of Forensic Medicine, socio-demographic characteristics from the 2005 census, location of police stations from the Metropolitan Police of Bogota, and presence of criminal groups and illegal markets from interviews with police precinct commanders. The research employed Principal Components Factor Analysis (PCFA) to create ecological constructs, and Exploratory Spatial Data Analysis (ESDA) and Spatial Regression Analysis (SRA) to examine patterns of spatial dependence in the outcome and predictor variables. Results provide partial support for social disorganization theory to the extent that concentrated disadvantage, social isolation, and residential mobility positively predict homicide rates above and beyond the effect of the presence of criminal groups and other controls. Only one proxy measure of the public level of control (presence of police) was significant, but its effect was in the opposite direction to what was hypothesized. However, this effect disappeared in the final model once the temporal lag of homicide rates was introduced. The study makes several contributions to the literature including testing the external and construct validity of social disorganization and systemic model of control measures, proposing a mixed-methods approach to get a more nuanced understanding of the spatial distribution of homicide rates, and suggesting policy implications to reduce the effects of disadvantage as potentially effective strategies in preventing violent crime at the neighborhood level. In sum, the study provides some evidence in favor of the usefulness of social disorganization theories to understand violent crime in Latin American cities. Replications in the region will be needed to assess the generalizability of these findings.


Digital reproduction from the UMI microform.

Included in

Criminology Commons