Pexel and heatmap visual analysis of multidimensional gun/homicide data

We present a visual analysis tool for mining correlations in county-level, multidimensional gun/homicide data. The tool uses 2D pexels, heatmaps, linked-views, dynamic queries and details-on-demand to analyze annual county-level data on firearm homicide rates and gun availability, as well as various socio-demographic measures. A statistical significance filter was implemented as a visual means to validate exploratory hypotheses. Results from expert evaluations indicate that our methods outperform typical graphical techniques used by statisticians, such as bar graphs, scatterplots and residual plots, to show spatial and temporal relationships. Our visualization has the potential to convey the impact of gun availability on firearm homicides to the public health arena and the general public.