Published on December 11, 2025
The analysis, Who Gets Arrested in America: Trends Across Four Decades, 1980-2024, reconstructs a continuous series of national arrest counts and rates from 1980 through 2024. To do so, it aligns the Office of Juvenile Justice and Delinquency Prevention’s (OJJDP) historical, weighted arrest estimates with contemporary data from the FBI’s Crime in the United States series. Using overlapping years (2005–2020), the analysis matches demographic shares from FBI arrest tables to OJJDP’s totals, applies those shares to the FBI’s weighted national counts, and produces arrest estimates by age, sex, race, and offense type. Population estimates from the Centers for Disease Control (CDC) and the U.S. Census Bureau were then used to generate arrest rates. This approach preserves historical comparability, restores demographic detail that has not been published by the federal government since 2020, and enables future annual updates as new FBI data become available.
Why This Analysis Was Needed
The FBI produces a national estimate of arrests that accounts for partial- and non-reporting jurisdictions annually (Table 29 of the online version of Crime in the United States1). These weighted arrest totals closely track the historical estimates published by OJJDP. Most years, they differ by less than 1%, with the differences primarily stemming from methodological differences in weighting procedures (See Figure S1.)
Figure S1. Arrest Counts by Offense Type and Data Source, 1980-2024
However, the FBI’s weighted national totals are not disaggregated by age, sex, race, or offense type. While demographic breakdowns are available in other FBI tables, those figures reflect only reported arrests and are not adjusted for non-reporting agencies. This results in undercounts and prevents accurate estimation of national rates (Figure S2).
Figure S2. Comparing Weighted Arrests to Reported Arrests, 1980-2024
For decades, OJJDP addressed this gap by publishing weighted national arrest counts and rates by age (juveniles aged 10–17 and adults aged 18 and older), sex, race, and offense type. But that series ended in 2020. To extend it, we developed a method to disaggregate the FBI’s weighted national arrest estimates by demographic group. This allowed for reconstruction of the full arrest series through 2024 and the restoration of detail that had disappeared from federal reporting.
Data Sources
This analysis draws on multiple federal datasets to reconstruct national arrest rates by age, sex, race, and offense type. Two primary arrest data sources were used to build the core series, supported by three population datasets for rate calculations (Table S1). Beginning in 2021, the FBI began publishing weighted national arrest estimates (counts and rates) on the Crime Data Explorer, based on law enforcement agencies’ regular reporting through the National Incident-Based Reporting System (NIBRS). These estimates represent an important contribution to national crime measurement and provide timely national indicators during the NIBRS transition period.
However, the NIBRS estimation series currently cannot serve as a substitute for the long-term trend reconstruction presented here. The NIBRS estimates include only index offenses (violent and property crime) and do not provide estimated arrests for drug or other offenses. In addition, they are not disaggregated by age, sex, or race, demographic dimensions that are central to understanding shifts in who is being arrested and for what crimes. Until the FBI’s NIBRS estimates offer full demographic detail across offense types, extending the historical trend requires the approach used here.
Table S1. Data Sources Used in the Analysis
Arrest counts were drawn from four FBI tables from the online version of Crime in the United States from 2005 to 2024.
- Table 29. National estimated arrests by offense type (not broken out by age or sex)
- Tables 39 and 40. Arrests by age, sex, and offense type (only for reporting agencies)
- Table 49. Arrests by age, race, and offense type (only for reporting agencies; not sex-specific)
As in OJJDP’s historical series, the word “juveniles” refers to those under age 18. Where possible, arrests of juveniles aged 10 to 17 were used. Table 49 aggregates all youth under 18, so excluding juveniles younger than 10 was not possible for race analyses.
A very small share of arrests involve children younger than age 10 (less than 0.5% annually). These arrests have declined steadily—from roughly 35,000 in the 1980s and 1990s to about 1,500 in 2024—and are excluded from the analysis.
Aligning Offense Categories Over Time
In 2013, the FBI replaced its legacy definition of rape—historically defined as “carnal knowledge of a female forcibly and against her will”—with a broader definition covering any non-consensual sexual act, regardless of the victim’s gender or use of force. This shift breaks the historical trend, and OJJDP adjusted its reporting in response:
- 1980–2012: OJJDP continued to report arrests for rape and “sex offenses (except rape and prostitution),” but folded both into an “other” category and defined violent crime as homicide, robbery, and aggravated assault only.
- 2013–2024: OJJDP stopped publishing arrest counts for rape and other sex offenses. Cells for those crimes appear as NA.
To maintain consistent offense categories from 1980 to 2024, this analysis:
- Used the Violent Crime Index (homicide, rape, robbery, and aggravated assault), wherever data were available.
- Placed arrests for other sex offenses in the “other” category wherever data were available.
Applying a uniform classification ensures that long-term trends are comparable before and after the FBI’s 2013 definition change. This approach adds roughly 20,000 annual arrests for rape to the violent crime category and about 40,000 annual arrests for other sex offenses to “other” in the post-2013 period. That adjustment makes violent crime estimations based on FBI data modestly higher than those used in OJJDP’s reported series beginning in 2013.
Property crime and drug offenses follow historical definitions (Property Crime Index and Drug Abuse Violations), with all remaining offenses grouped into the “other” category.
Methodology
Generating Weighted Arrest Counts by Demographic Group
To rebuild national arrest counts disaggregated by age, sex, race, and offense type, this analysis uses FBI arrest tables to allocate the FBI’s national estimated totals across demographic groups. The summary below outlines how counts were generated and validated against OJJDP’s historical weighted series.
Estimating Arrests by Age
First, arrest estimates were calculated by age group (juveniles ages 10–17 and adults 18+). FBI Tables 39 and 40 report arrests by age and sex for agencies that submitted complete annual data. These figures needed to be adjusted to align with the FBI’s national weighted estimates in Table 29, which account for partial- and non-reporting jurisdictions. To accomplish that, arrest shares were calculated separately for adult males, adult females, juvenile males, and juvenile females by offense type and applied to the FBI’s national totals. Breaking out arrest by age, sex, and offense type resulted in the closest possible estimates of total arrests, arrests by age, arrests by age and sex, and arrests by age, sex, and offense type. Arrests of children under age 10 were excluded.
For example, in 2005, there were 445,204 reported violent crime arrests of people aged 10 and older in UCR reporting jurisdictions: 308,503 adult males, 66,861 adult females, 56,967 juvenile males, and 12,873 juvenile females. Adult males represented 69.3% of violent arrests, adult females 15.0%, juvenile males 12.8%, and juvenile females 2.9%.
The FBI estimated that, after accounting for partial and non-reporting jurisdictions, there were 603,503 total violent arrests in 2005. Applying the calculated group shares (using five decimal places) to that total yielded an estimated 418,196 violent arrests of adult males, 90,634 of adult females, 77,222 of juvenile males, and 17,450 of juvenile females. This same procedure was repeated for property, drug, and other arrests. Totals were then aggregated to generate adult and juvenile estimates.
Comparisons with OJJDP’s weighted estimates show close alignment across overlapping years, with juvenile estimates slightly higher before converging by 2010 (Figure S3).
Figure S3. Estimated Arrests by Age and Data Source, 1980-2024
Estimating Arrests by Age, Sex, and Offense Type
Next, the same method was used to generate arrest estimates by age, sex, and offense category. Arrest shares by age and offense type were calculated from FBI Tables 39 and 40 and applied to national weighted arrest totals by offense type (Table 29). This produced national estimates for adult males, adult females, juvenile males, and juvenile females for each offense category (Figures S4–S7).
As expected, adult and juvenile violent crime estimates in this series are modestly higher than OJJDP’s beginning in 2013 because the weighted FBI data include rape in the violent crime category. The difference is most pronounced for males. A second discrepancy appears in the “other” offense category for juveniles from 2005 to 2009. The source of this divergence is unclear, particularly because the pattern does not appear in the adult data. The two juvenile series converge beginning in 2010.
Figure S4. Estimated Adult Male Arrests by Offense Type and Data Source, 1980-2024
Figure S5. Estimated Adult Female Arrests by Offense Type and Data Source, 1980-2024
Figure S6. Estimated Juvenile Male Arrests by Offense Type and Data Source, 1980-2024
Figure S7. Estimated Juvenile Female Arrests by Offense Type and Data Source, 1980-2024
Estimating Arrests by Age and Race
To allocate national arrest estimates by race, arrest shares were drawn from Tables 49B (juveniles) and 49C (adults) in the FBI’s online version of Crime in the United States, which report arrests by age and race for agencies submitting a full year of data. These tables do not report arrests by race and sex. Juvenile tables group all individuals under 18 together, so juvenile estimates here reflect arrests of youth under age 18 rather than the 10–17 age range used elsewhere.
Arrests are reported across four race groups: White, Black, American Indian/Alaska Native (AI/AN), and Asian. Hispanic ethnicity is reported separately and cannot be integrated into the race series without double-counting, because race totals already sum to 100%. Beginning in 2013, the FBI split the Asian category into Asian and Native Hawaiian/Other Pacific Islander; these categories were merged in this analysis to maintain consistency across years.
Not all arrests include race information. The share of arrests with missing race data is small but non-trivial—fewer than 200,000 adult arrests and 20,000 juvenile arrests per year—and rose notably in 2022, potentially due to uncertainty around coding multiracial identities or other unknown factors (Table S2). To address this, race shares were calculated only among arrests with valid race information and applied to the FBI’s national arrest totals by offense and year.
Table S2. Number and Share of Reported Arrests with Race Missing by Age Group, 1980-2024
For each year, arrests by offense type were allocated across White, Black, AI/AN, and Asian groups, and were summed separately for adults and juveniles using race-specific shares. The proportion of arrests attributable to each racial group was then calculated by dividing each subtotal by the total arrests for that offense category. Because the racial distribution of arrests varies by offense, these proportions were used to more accurately allocate the national totals across racial groups. Because OJJDP did not publish race-by-offense estimates, resulting race-specific arrest totals were aggregated to the same race-by-age groupings used in OJJDP’s series to maintain historical continuity.
The reconstructed adult series closely tracks OJJDP’s estimates, with small overestimates for Black and AI/AN adults and slight underestimates for Asian adults (Figure S8). Notably, discrepancies exist for AI/AN adults in 2018, 2019, and 2020, as well as for Asian adults in 2007, 2011, and 2012. It is unclear whether these discrepancies are due to within- or across-race weighting applied by OJJDP, the number of arrests with missing race information, a tendency for certain racial groups to be over- or under-represented in jurisdictions that report arrests to FBI, or other factors.
Figure S8. Estimated Adult Arrests by Race and Data Source, 1980-2024
The resulting juvenile estimates show somewhat greater discrepancies than those for adults (Figure S9). Juvenile arrests generated from the FBI data tend to overestimate White, Black, and AI/AN arrests between 2005 and 2009, but converge with the historical OJJDP series beginning in 2010. Estimates for Asian juveniles do not align until 2013, coinciding with the FBI’s change in the Asian race categorization that year.
Figure S9. Estimated Juvenile Arrests by Race and Data Source, 1980-2024
Generating Weighted Arrest Rates by Demographic Group
Arrest rates were calculated using age- and sex-specific population estimates. To maintain consistency with the historical OJJDP series, population denominators from the CDC’s Bridged-Race Population Estimates were used through 2020, as this dataset provides annual counts by age, sex, and race that align with FBI reporting categories. Because these data are no longer available after 2020, rates for 2022 to 2024 were produced using population estimates from two sources: the U.S. Census Bureau’s American Community Survey 1-Year Estimates for age- and sex-specific denominators, and the CDC’s Population Estimates Program for race-specific denominators. The Population Estimates Program was used instead of the American Community Survey for race because its categories match the FBI’s reporting structure, while the American Community Survey race categories are more detailed and not directly comparable. Rates are expressed per 100,000 population. No rates were produced for 2021 due to the FBI’s transition to a new reporting system, which created a one-year break in national estimates.
Arrests by Age and Sex
Arrest rates generated from FBI data closely track OJJDP’s historical series across the 2005 to 2020 overlap period (Figure S10). For adults, the two sources produce nearly identical trends for both men and women. For juveniles, FBI-based rates run slightly higher than OJJDP’s from 2005 to 2009, but the series converge by 2010 and move in tandem thereafter.
Figure S10. Estimated Arrest Rates by Age, Sex, and Data Source, 1980-2024
Arrests by Age, Sex, and Offense Type
By offense category, FBI-based adult arrest rates track OJJDP’s historical estimates extremely closely (Figure S11). Violent, property, drug, and “other” offense trends align across the overlap period, with only minor deviations—most noticeably small gaps for men in 2013 and 2019. These results support the continuity and comparability of the reconstructed series.
Figure S11. Estimated Adult Arrest Rates by Age, Sex, Offense Type, and Data Source, 1980-2024
For juveniles, offense-specific patterns follow the same overall alignment (Figure S12). Rates based on FBI data run somewhat higher than OJJDP’s between 2005 and 2009—particularly for violent, property, and “other” offenses—before converging around 2010. The only persistent difference is a modest gap in juvenile male arrests for violent crime, for which FBI-based estimates remain slightly elevated across overlap years.
Figure S12. Juvenile Arrest Rates by Age, Sex, Offense Type, and Data Source, 1980-2024
Arrests by Age and Race
Reconstructed adult arrest rates align closely with OJJDP’s historical estimates across most racial groups (Figure S13). Small differences appear for AI/AN and Asian adults, whose rates converge with OJJDP’s series in 2011 and 2013, respectively. The earlier convergence for AI/AN adults likely reflects adjustments to population estimates following the 2010 Census, while the later alignment for Asian adults appears tied to differences in how arrests were classified and reported across the two series.
Juvenile arrest rates follow a similar pattern. Rates for White, Black, and AI/AN youth converge with OJJDP’s estimates around 2011, with Asian youth aligning slightly later in 2013. This coincides with the FBI change that separated Asian and Native Hawaiian/Other Pacific Islander categories.
Figure S13. Arrests by Age, Race, and Data Source, 1980-2024
Final Assembly and Ongoing Updates
To create a single, continuous series, this analysis uses OJJDP’s weighted arrest estimates through 2012 and transitions to the FBI-based series in 2013, the year the two sources converge across age, sex, race, and offense type. This transition point also ensures that arrests for rape and other sex offenses are consistently incorporated throughout the entire 1980 to 2024 period, following the FBI’s definitional change in 2013. Arrests of children younger than age 10 are excluded across all years; these cases represent less than 0.5% of total arrests.
The resulting dataset preserves the structure of OJJDP’s historical series, restores demographic detail last published federally in 2020, and enables annual updates using the FBI’s ongoing Crime in the United States reports. All steps and decisions are documented to support transparency and replication going forward.
Descriptions of the final trends are available here. As new data are released from the FBI, arrest trends will be periodically updated and maintained within CCJ’s data library, The Footprint: Tracking the Size of the American Criminal Justice System.
