This article is available as a YouTube podcast.
Your jurisdiction’s crime rates could increase or decrease significantly.
It will have vast implications for police and justice professionals.
CrimeinAmerica.Net-Chat GPT’s “Top 10 Sources for Crime in America” based on primary statistical sources with trusted secondary analysis.
Author
Leonard Adam Sipes, Jr.
Former Senior Specialist for Crime Prevention and Statistics for the Department of Justice’s clearinghouse. Former Director of Information Services, National Crime Prevention Council. Former Adjunct Associate Professor of Criminology and Public Affairs-University of Maryland, University College. Former police officer. Retired federal senior spokesperson.
Former advisor to presidential and gubernatorial campaigns. Former advisor to the “McGruff-Take a Bite Out of Crime” national media campaign. Produced successful state anti-crime media campaigns.
Thirty-five years of directing award-winning (50+) public relations for national and state criminal justice agencies. Interviewed thousands of times by every national news outlet, often with a focus on crime statistics and research. Created the first state and federal podcasting series. Produced a unique and emulated style of government proactive public relations.
Certificate of Advanced Study-The Johns Hopkins University.
Author of ”Success With The Media: Everything You Need To Survive Reporters and Your Organization,” available at Amazon and additional bookstores.
Crime in America.Net-“Trusted Crime Data, Made Clear.”
Quoted by The Associated Press, USA Today, A&E Television, Inside Edition Television, Oxygen and allied publications, The nationally syndicated Armstrong Williams Television Show (30 times), Department of Justice documents, multiple US Supreme Court briefs, C-SPAN, the National Institute of Health, college and university online libraries, multiple books and journal articles, The Baltimore Sun, The Capital Gazette, MSN, AOL, Yahoo, The Daily Beast, The Huffington Post, JAMA, News Break, The National Institute of Corrections, The Office of Juvenile Justice And Delinquency Prevention, The Bureau of Justice Assistance, Gartner Consulting, The Maryland Crime Victims Resource Center, Law.Com, The Marshall Project, The Heritage Foundation via Congressional testimony, Law Enforcement Today, Law Officer.Com, Blue Magazine, Citizens Behind The Badge, Police 1, American Peace Officer, Corections.Com, Prison Legal News, The Hill (newspaper of Congress), the Journal of Offender Monitoring, , Yomiuri Shimbun (Asia’s largest newspaper), LeFigaro (France’s oldest newspaper), The Readers Digest (plus international editions), Money Geek, Forbes, Newsweek, The Economist, The Civic Research Institute, The Toronto Sun, Homeland Security Digital Library, The ABA Journal, The Cambridge University Press, The Daily Express (UK), The Harvard Political Review, The Millennial Source, The Federalist Society, Lifewire, The Beccaria Portal On Crime (Europe), The European Journal of Criminology, American Focus and many additional publications.
Sign up for notice of new articles on the front page of this site.
A comprehensive overview of crime for recent years is available at Violent and Property Crime Rates In The U.S.
Quotes
Crime stats don’t lie-but they don’t tell the whole truth either. Crime statistics are the wild west of sociology. Based on current USDOJ data, anyone can make any case they want as to crime in America. That’s about to change.
Background
The Marshall Project: “More money to collect crime data. Trump’s order pledged to spend more money on collecting and standardizing crime data. Crime statistics became the source of a clash between Trump and law enforcement leaders during his 2024 campaign, with some police chiefs disputing Trump’s claims that crime was “out of control” in cities across America.”
The Marshall Project: Monday’s executive order on policing included a mandate to “increase the investment in and collection, distribution, and uniformity of crime data across jurisdictions.” Project 2025 — a conservative policy blueprint for the Trump presidency — wrote favorably of the Justice Department’s National Crime Victimization Survey, saying officials “should prioritize and sufficiently fund it.”
President Trump’s Executive Order
The President ordered the Office of Justice Programs and the Bureau of Justice Statistics to examine the feasibility of combining the USDOJs National Crime Victimization Survey with reported crime data produced by the FBI.
NCVS data would be the primary set of numbers, supplemented by FBI data.
The Department of Justice and BJS-funded researchers have begun testing a method that could change how we measure crime — and therefore how policymakers and the public understand it.”
Notes
Throughout this article I make reference to a document funded by the Bureau of Justice Statistics as offered by the Iowa State University. I refer to the research as an BJS document.
Article
This article addresses the President’s executive order and how crime statistics via the US Department of Justice could (will?) change.
Fustration
When I became the senior specialist for crime prevention at the US Department of Justice’s clearinghouse, the National Criminal Justice Reference Service, I became immersed in crime statistics, including those provided by the Bureau of Justice Statistics’ National Crime Victimization Survey and reported crimes as compiled by the FBI.
The frustration within the USDOJ’s Office of Justice Programs, the overall agency for federal crime research and funding, was that the lessons offered by the National Crime Victimization Survey were being ignored. It’s still happening today, and few (especially within the media) know of the 50-year-old NCVS.
Not only did the NCVS provide MUCH larger crime numbers than the FBI data, but it also provided a better quality of information to enhance our understanding of crime. From the NCVS, I could tell you who is victimized, where they live, the kind of housing they occupy, the age, sex, and income of victims, what happens when you fight back against violent criminals, where crime happens, the race or ethnic background of criminals and victims, and endless other variables.
Per the Bureau of Justice Statistics of the USDOJ, when you consider that 30 percent of property crimes are reported to law enforcement (approximately 70 percent of what we call crime are property events), and about half of violent crimes are reported, and that 38 percent of urban violent crimes are reported, you come to realize that we within the justice system and the public are not getting the best possible understanding of crime in America.
Unreported crimes are just one example of the inadequacies of relying solely on reported crimes to inform crime policy. There are many more.
Reported crimes have declined nationally by 3-4 percent per the FBI, but according to the NCVS, there was a 44 percent increase in violent crime rates in 2022, and these rates have remained stable for 2023 and the latest recent report for 2024.
The FBI’s National Incident-Based Reporting System
The FBI’s National Incident-Based Reporting System attempts to rectify the inadequacies of our understanding of reported crime by collecting more qualitative data and recording the number of individual crimes per incident. It’s a much-needed improvement, but accessing and understanding that data through the FBI’s website is a challenge.
The Report
A new Bureau of Justice Statistics-funded report from the Iowa State University explores whether the nation can finally move beyond relying almost entirely on crimes reported to police to better understand crime in America.
Any analysis, conclusions, or opinions expressed in the report do not necessarily represent the views, opinions, or policies of BJS or the U.S. Department of Justice (standard language for USDOJ-funded research).
Police reports miss a large share of violent and property crime because most victims never contact law enforcement. The report tests a new statistical approach that blends the USDOJ’s National Crime Victimization Survey—which captures both reported and unreported crime—with the FBI’s Uniform Crime Reports, which exist for every state.
The goal is to create reliable state-level crime estimates that reflect the full picture of victimization, not just what ends up in police databases. The study is a feasibility test, but if the approach is refined and adopted, it could give criminologists, policymakers, and the public a much more accurate and realistic understanding of crime across the country.
Note that the Bureau of Justice Statistics has been examining an expansion of the National Crime Victimization Survey to states for several years. For those interested, redesigns of the National Crime Victimization Survey have been ongoing; see the website for the Bureau of Justice Statistics.
What This Means for Police, Criminology And The Justice System
The President’s Commission on Law Enforcement and Administration of Justice in the 1960s (plus other federal crime commissions) promised us a new understanding of crime, criminals, and crime prevention that never fully materialized.
The new data has the potential to revolutionize our understanding of the justice system and the best methods to combat crime.
It also means:
That your jurisdiction’s crime rates could increase or decrease significantly.
The guiding document says it will offer state crime rates, but it “may” eventually offer crime rates for cities and metropolitan areas along with state data, depending on sample size. Washington, D.C., is included in the Iowa State University document. If D.C. is included, then other cities or metropolitan areas may have sufficient data based on the combination of NCVS and NIBRS data.
Combining the National Crime Victimization Survey with the National Incident-Based Reporting System (reported crime) will provide us with an endless array of new understandings about crime. Simply put, larger numbers provide better results. They reduce statistical uncertainty and allow subnational (state-local) estimates.
For example, we know that the majority of violent crime happens between people who know each other, and we know that the majority of rapes and sexual assaults happen in residences. Thus, we can tell women that they need to be careful who they invite into their homes or whose homes they enter.
We can see the possible interplay between cities, metropolitan areas, and states to see if crime patterns are different depending on location. Violent crimes could explode in one jurisdiction. Then we get the word out via social and conventional media as to the risks potential victims need to be aware of.
What’s The Problem BJS Is Trying To Solve?
Because the NCVS was originally designed for national estimates, it doesn’t always have enough respondents in every state (especially smaller ones) to produce reliable state-level crime rates.
That gap limits researchers, policymakers, and journalists — we end up largely relying on FBI data, which reflects only reported crime. That undercounts the “true” amount of crime overall.
What The New BJS Study Does—And Proposes As A Potential Solution
The report uses a statistical modeling approach — specifically a Bayesian multivariate lognormal model — to blend NCVS and UCR data. In effect, it tries to use the strengths of both: the reality-based victimization data from NCVS, plus the coverage and ubiquity of UCR data (which exists for all states).
The goal: produce state-level estimates of crime victimization — including both reported and unreported crime — for all 50 states plus D.C., even where NCVS alone lacks enough data.
Why This Matters—The Potential Benefits
If the method works (and gets refined), it could represent a major advance for policing, criminology, policy, media reporting, and public understanding:
More accurate crime rates: Because many crimes go unreported, relying on police data alone undercounts total crime. This blended approach could capture a fuller picture of crime — including the “dark figure” of unreported crime.
State-level granularity: Policymakers and researchers could see meaningful crime victimization estimates for every state — not just national aggregates — enabling better resource allocation, policy evaluation, and comparisons.
Better-informed public debate: More realistic crime data could help resolve some of the conflicting narratives: “Crime is down” (based on police reports) vs. “real victimization is up” per the NCVS.
More useful for social science research: With better data, criminologists could more reliably study the causes and patterns of crime — linking victimization rates to social, economic, demographic, or policy variables.
This Is A Proposal And Not A Green Light For Media Or Policy
BJS itself calls the Iowa State University/BJS report a feasibility study, not a release of new official crime statistics.
That means:
These blended numbers in the report are experimental estimates, not “official” data. They should be treated as what might be possible, not what is yet proven or final.
The method remains under evaluation — there are important technical challenges to overcome (sampling limitations, statistical uncertainty, inference assumptions).
BJS is not yet endorsing these estimates for public policy, media citation, or official state-by-state comparisons.
What Cops, Criminologists, Policymakers, Journalists Should Know
This new BJS report doesn’t deliver a new official crime dataset — yet. Rather, it offers an important proof-of-concept: that it may be (probably is) possible to combine victimization surveys and police data to generate realistic, state-level crime estimates that include both reported and unreported crime.
For now, the work remains exploratory. But it offers significant hope that we might soon see a far better, more honest “dashboard of crime” for the entire country — not dependent solely on what gets reported to police.
Conclusions
The report calls for the possibility of offering combined NCVS and FBI data over six years. This is not unusual for Bureau of Justice Statistics data, which often uses three years on the basis that the larger the numbers, the more accurate the findings. BJS has the capability now of producing NCVS direct estimates for the 22 largest states based on a 3-year average. These direct estimates could naturally be paired for all states based on a 6-year average.
However, it’s my opinion that President Trump likes policies to unravel quickly, so it wouldn’t surprise me that a report combining NCVS and FBI data comes out during his term in office.
It’s also my opinion that we will see combined FBI and NCVS data for large cities and metropolitan areas before Trump leaves office.
Appendix-An Analogy
Imagine you want to estimate how many people in each town got sick during flu season — but you don’t have direct data for every town (the survey only covers some). But you do have data on how many people visited hospitals in each town (hospital-visit records). So you build a model that says: “If hospital visits were high, maybe sickness was high; if hospital visits were low, maybe sickness was low.”
That works — if hospital visits reliably reflect actual sickness. But if a lot of people with flu don’t go to hospitals, or hospital visits reflect other illnesses too, then hospital-visit data become a “weak predictor” of true sickness rates. Your model’s estimates for towns then become shaky.
In this case: the “hospital visits” = UCR crime reports; the “true sickness” = actual victimization as captured by NCVS. Because the UCR crime reports don’t reliably track real victimization (for many crime types/states), the model’s estimates can be weak or unstable.
This article is a combination of my original writing supplemented by ChatGPT’s analysis of the research.
One Reply to “America’s Crime Statistics Will Change—Here’s Why”