This application enables users to identify similar geographic areas based on a set of covariates (characteristics) related to each area.
In crime research, accurately identifying similar treatment and control areas is crucial for evaluating the effectiveness of interventions, policies, or prevention strategies. By ensuring that comparison areas share similar characteristics, researchers can more confidently attribute observed differences in outcomes—such as reductions in crime rates or changes in behavior—to the intervention rather than underlying differences between areas. Without careful matching, results may be skewed by confounding variables, leading to incorrect conclusions. This approach allows for a more rigorous evaluation of crime-related interventions, helping policymakers and practitioners to design and implement more effective strategies for crime reduction and public safety.
This application enables users to identify similar geographic areas based on a set of covariates (characteristics) related to each area. Two methods have been implemented enabling the user to both identify similar areas based on a set of characteristics (Nearest Neighbour Matching) but to also identify similar areas once a treatment has already been assigned (Propensite Score Matching).
View Project on GitHub Star Follow @mopac-ds
Screenshots of Our Application are Shown Below
Application Overview:
Area Matching:
Control-Treatment Matching:
Quality Assurance:
