Because of numerous problems with this data (http://trendct.org/2016/11/21/connecticut-saw-a-drop-in-reports-of-hate-crime-incidents-in-2015/) naive year-to-year comparisons of total hate-crime incidents are not particularly informative, and at worst misleading. They don't take into account that the number of police agencies participating changes from year to year, and that the list of agencies participating could be completely different. If high-hate-crime jurisdictions go on and off the list, the appearance of significant swings could result.
- per-capita analysis in cities.ipynb - notebook used to adjust data for population.
- hate crimes - table 12.ipynb - notebook to analyze table 12, a simple state-by-state incident count for all six year
- data/*.xls - raw spreadsheets, unaltered from FBI site (https://www.fbi.gov/investigate/civil-rights/hate-crimes)