Human Trafficking Colombia 2012-2023
Description
Since you need to fill this into a text box, it should be a single, cohesive narrative that flows logically. I have condensed the information into a professional paragraph that answers all the requirements: hypothesis, findings, interpretation, and data origin. Data Description: The research hypothesis for this study posited that Colombia’s National Strategy to Combat Human Trafficking (2016–2020) would be associated with a measurable change in the trends and spatial distribution of reported trafficking victimization, but that its effectiveness would be moderated by local socioeconomic conditions and institutional presence. Our data, comprised of municipal-level victimization records and socioeconomic indicators, shows significant spatial clustering of trafficking reports, particularly in transit corridors and regions with persistent conflict. The analysis reveals that while the national strategy coincided with changes in reporting patterns, these effects were not uniform across the country, suggesting that trafficking risk often shifted rather than disappeared. Notable findings from the negative binomial regression and spatial hotspot analyses indicate that factors such as informal labor markets and local poverty levels are stronger predictors of trafficking risk than the presence of the policy alone. This suggests that the data should be interpreted as a call for place-based, data-driven interventions rather than centralized, "one-size-fits-all" strategies. The dataset was constructed by merging official victimization counts from National Police with municipal demographic data from DANE. To enable replication, the data is organized by municipal DANE codes and has been cleaned to ensure longitudinal consistency across the 2016–2023 period.
Files
Steps to reproduce
The final dataset was constructed through a multi-stage geospatial and statistical workflow designed to ensure high replicability and data integrity. Raw victimization records were initially sourced from the Policía Nacional (SIEDCO), covering the period from 2015 to 2023. These records were cleaned in Excel where duplicates were removed using unique case identifiers and missing geographic values were imputed via municipal cross-referencing. To analyze socio-economic correlations, these data were harmonized with census indicators from DANE using the municipal coding system as the relational key. Spatial analysis was performed using the ARCGIS to calculate Global and Local Moran’s I, identifying significant clusters of victimization. All spatial joins were executed via a Point-in-Polygon protocol to aggregate individual incidents to the municipal level while maintaining victim anonymity. This standardized workflow, documented in the accompanying replication scripts, ensures that the spatial patterns identified are statistically robust and consistent with established Colombian crime-mapping methodologies