Frequently Asked Questions

Taking a sample of cities is a way to understand trends without having to wait months (or more) for formal crime estimates to be released, and more cities of data makes for better predictive power. Murder data for every city over 250,000 (about 90 agencies) between 2010 and 2020 was within 2.5 percentage points of the formal national change in murder over that span while data for every city over 100,000 (about 300 agencies) was within 1.5 percentage points. Data for every city of 50,000 or more (about 850 agencies) should be within roughly 1 percentage point of the national change on average.

Data is updated roughly a month and a half after the end of the reporting month, so data through June is available in mid-August. This delay is necessary because some agencies update a previous month’s data when reporting the most recent completed month. An agency might send its data for July to its state UCR program in early August along with information on a small but meaningful percentage of crimes that occurred in June. Reporting data any earlier would create a constant undercount of the most recently reported data.

There are many reasons why the figures reported for an agency may not exactly match what is reported in the Crime Index. Some of those reasons include: some incidents may be reclassified as a different offense category after it is published in their Records Management System, an agency may report incorrectly aggregated data on their website, or a few agencies may not report rape data publicly which would impact that agencies violent crime totals. The purpose of the Crime Index is to accurately portray trends, so minor inconsistencies between sources is an inevitable downside to working with imperfect data.

Data and graphics derived directly from the Crime Index should credit the Crime Index and the agency or state UCR program that produced the data as its source through the timeframe of available data at that time. Data that is included in the Crime Index that users take directly from an agency or state UCR program’s original source can be credited solely to that agency or state UCR program.

Good question!

The Crime Index is funded through a generous grant from Arnold Ventures. We have reached a number of our original goals for sample size and coverage, so next steps will likely include subject matter expansion. Of particular interest is expanding to visualizing trends for crime types that are best sourced from NIBRS. This includes crimes like carjackings as well as fatal and non-fatal shootings.