A crime rate is calculated by dividing the number of reported offenses by the population of the area, then multiplying by 100,000. That produces the standard figure you see quoted in news stories and government reports: offenses per 100,000 inhabitants. The formula is simple. What feeds into it, and what gets left out, is where the real work is.
(Number of Reported Crimes ÷ Population) × 100,000 = Crime Rate per 100,000 Inhabitants
A Worked Example
Take a city of 250,000 residents that records 1,200 violent crimes in a year. The violent crime rate is (1,200 ÷ 250,000) × 100,000, which comes to 480 per 100,000. The multiplication by 100,000 is a scaling convention. Without it you’d be comparing decimals like 0.0048 and 0.0031, which is technically accurate but useless at a glance.
The Bureau of Justice Statistics uses a different scaling factor for its victimization data, expressing rates per 1,000 people age 12 or older rather than per 100,000. The 2024 National Crime Victimization Survey reported a violent victimization rate of 23.3 per 1,000.1Bureau of Justice Statistics. Criminal Victimization, 2024 Same kind of calculation, different unit, which is why NCVS numbers look nothing like FBI numbers even when they describe the same country.
Where the Population Number Comes From
The denominator is not simply the U.S. population. The FBI uses the aggregated populations covered by the law enforcement agencies that submitted a full 12 months of data that year. Those figures come from U.S. Census Bureau decennial counts and annual estimates, with the FBI computing individual growth rates for every city and county to project current-year numbers.2Federal Bureau of Investigation. Methodology
The practical effect is that the denominator moves year to year based on which agencies participated. If a large city’s police department fails to submit data, both its crimes and its population drop out of the national calculation. Reporting to the FBI’s Uniform Crime Reporting Program is voluntary, and although more than 18,000 agencies submit data, coverage has never been complete.3Federal Bureau of Investigation. Crime/Law Enforcement Stats (UCR Program) As of mid-2024, agencies covering roughly 82 percent of the U.S. population were actively reporting through the National Incident-Based Reporting System.4Bureau of Justice Statistics. National Incident-Based Reporting System (NIBRS) The remaining fifth is estimated, not directly counted.
Which Crimes Are in the Numerator
Not every offense feeds into a published crime rate. The FBI’s system has long sorted crimes into two tiers, and headline numbers almost always draw from the first.
Part I Offenses
Part I offenses are the serious crimes most likely to be reported and most consistently defined across jurisdictions. Violent Part I crimes are murder, rape, robbery, and aggravated assault. Property Part I crimes are burglary, larceny-theft, motor vehicle theft, and arson. Human trafficking, covering both commercial sex acts and involuntary servitude, was added more recently.5Federal Bureau of Investigation. Offense Definitions When you see a “violent crime rate” or “property crime rate” quoted, it refers to these.
Part II Offenses
Part II covers everything else: simple assault, fraud, vandalism, drug offenses, DUI, weapons violations, disorderly conduct, and more. These are tracked primarily through arrest data rather than incident reports, so they rarely appear in the crime rate figures a city or state publicizes.
Violent and Property Rates Are Reported Separately
The FBI does not publish a single blended crime rate. Violent and property crimes each get their own rate, because property crime is far more common and blending the two would mask what is happening with violence.5Federal Bureau of Investigation. Offense Definitions A city can see its combined rate drop while murders and assaults climb, if auto thefts fell enough to offset the increase. The first question to ask about any crime rate figure is whether it covers violent crime, property crime, or a specific offense.
The Numerator Only Counts Reported Crime
The formula’s biggest limitation is baked into the numerator. The FBI counts crimes that someone told police about. Anything unreported is invisible.
To fill that gap, the Bureau of Justice Statistics runs a completely separate measurement system. The National Crime Victimization Survey interviews roughly 240,000 people in about 150,000 households each year, asking whether they experienced crime in the prior six months, whether or not they told police.6Bureau of Justice Statistics. National Crime Victimization Survey (NCVS) Because it is a survey, results are weighted to match known population totals and to compensate for non-response.
The two systems frequently point in different directions, and neither is wrong. Changes in whether people trust police can move the FBI rate up or down without any change in actual crime, because the call to police is what creates the data point.7Bureau of Justice Statistics. True Crime Stories? Accounting for Differences in Our National Crime Indicators The NCVS avoids that problem but has its own: respondents forget incidents, decline to mention crimes committed by family members, and collapse multiple similar victimizations into one report.
What Can Move the Rate Without More Crime Happening
Because the formula depends on reported offenses, several changes can push the rate up or down without any real change in criminal activity.
Reporting practices matter. A department that encourages victims to file reports and makes the process easy will produce higher counts than one where victims face long waits and bureaucratic friction. A rate increase sometimes reflects better reporting, not more crime.
Definitions change. When the FBI expanded its definition of rape in 2013, jurisdictions that adopted the new definition immediately showed higher rape counts. The measuring stick changed, not the underlying behavior.
Counting rules change too. Under the FBI’s old Summary Reporting System, the “hierarchy rule” required agencies to report only the most serious offense in any incident involving multiple crimes.8Federal Bureau of Investigation. Frequently Asked Questions About NIBRS Individual Agency Data A single incident involving a robbery, an aggravated assault, and a motor vehicle theft was counted as one robbery; the other offenses vanished from the data. In 2021 the FBI retired the summary system and made NIBRS the sole standard, which captures up to 10 offenses per incident.3Federal Bureau of Investigation. Crime/Law Enforcement Stats (UCR Program) A study of 2014 NIBRS data found that converting the detailed records back to summary format eliminated data from 10.6 percent of multi-offense incidents and reduced total reported crime by 2.1 percent. Current numbers aren’t necessarily higher because crime rose; the tool got better at catching what was already there.
Population fluctuations distort the denominator. A tourist city or college town has a small resident population but a much larger effective population committing and experiencing crimes. The resulting rate looks inflated compared with a stable suburb of the same size.
Where to Look Up the Numbers
The FBI publishes crime data through its Crime Data Explorer at cde.ucr.cjis.gov, with interactive tools, downloadable datasets, and filtering by agency, state, or offense. The Bureau of Justice Statistics publishes NCVS results at bjs.ojp.gov. Both are free. When you pull a figure from either, check the year, which agencies reported, and whether the rate covers violent crime, property crime, or a specific offense. A number without that context is just a number.