Download the US Crime Index dataset
The full dataset behind every page in our Crime Index: 759 US cities, 2025 data, with per-offense counts and a multi-year rate series for each city. Free to download and free to republish with attribution.
License & attribution
Published under CC BY 4.0. You may republish, chart and build on this data — including commercially — as long as you credit the source. Please attribute as:
HireSecurityNow.com US Crime Index — https://www.hiresecuritynow.com/crime-rate/methodology
Underlying incident data comes from FBI Crime Data Explorer (Summarized UCR/NIBRS) + U.S. Census ACS 5-year and is public record. The index, grading, and trend series are our derived work.
What's in the file
One row per city, 27 columns. Notes on the ones that need them:
| Column | Notes |
|---|---|
| city | |
| city_slug | |
| state | |
| state_abbr | |
| population | |
| crime_index | 0–100 composite where HIGHER is safer. See the methodology. |
| grade | A–F band derived from crime_index. |
| violent_rate_per_100k | |
| property_rate_per_100k | |
| homicide_rate_per_100k | |
| total_rate_per_100k | |
| count_homicide | |
| count_rape | |
| count_robbery | |
| count_aggravatedAssault | |
| count_burglary | |
| count_larceny | |
| count_motorVehicleTheft | |
| count_arson | |
| population_coverage_pct | Share of the city population covered by reporting agencies. |
| agency_reporting_pct | |
| data_limited | true when agency reporting coverage is too thin to treat the row as comparable. Filter these out for rankings. |
| index_reliable | |
| trend_reliable | |
| first_year | |
| last_year | |
| total_rate_change_pct | Percent change in total_rate_per_100k between first_year and last_year. Negative means crime fell. |
Before you publish a ranking
- Drop rows where
data_limitedis true — reporting coverage there is too thin to compare fairly. - Also drop rows where
series_plausibleis false. Those cities have a partial reporting year at one end of their series: the agency filed, so the coverage flags read clean, but only part of the year is in the numbers. It inflates the index and it is the single easiest way to publish a wrong ranking from this file. - Apply a population floor. Small cities swing enormously on a handful of incidents; our own trend ranking uses a 50,000 minimum for this reason.
- Reported crime is not all crime. Offenses go unreported at rates that differ by offense type and by city — the methodology page covers this "dark figure" limitation directly.
- Text columns starting with
= + - @carry a leading apostrophe so spreadsheets don't execute them as formulas. Numeric columns are never guarded this way, so they parse directly.
Working on a story?
We can pull a custom cut — a single metro, neighborhood-level breakdowns for the cities we have portal data for, or the state-by-state security licensing dataset. See the press page for contact details and ready-to-use figures.