Contents

• Federal Hate Crime Data Collection
• Caveats and Limitations to Federal Hate Crime Data
• UCR Hate Crime Data
• NCVS Hate Crime Data
• Findings from UCR and NCVS Hate Crime Data
• UCR and NCVS Victimizations, 2015-2024
• Overall Trends from 2015 to 2024
• Victim Characteristics
• Victim Age
• Victim Sex
• Victim Race and Ethnicity
• Incident Characteristics
• Offender Characteristics
• Proportionality of Hate Crime Victimizations by Race/Ethnicity and Religion Compared to the U.S. Population
• Policy Considerations for Congress

Summary

The Federal Bureau of Investigation (FBI) started its hate crime statistics program pursuant to the Hate Crime Statistics Act of 1990 (HCSA; P.L. 101-275). The HCSA directs the U.S. Department of Justice (DOJ) to acquire and report data on crimes that "manifest evidence of prejudice based on race, gender and gender identity, religion, disability, sexual orientation, or ethnicity, including where appropriate the crimes of murder, non-negligent manslaughter; forcible rape; aggravated assault, simple assault, intimidation; arson; and destruction, damage or vandalism of property." These data are collected by the local, state, federal, and tribal law enforcement agencies to which the offenses were reported and submitted to the FBI's Uniform Crime Reporting (UCR) program.

Hate Crime Victimization Rates by Data Source, 2015-2024

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: The NCVS estimates are based on the non-institutionalized U.S. resident population age 12 or older. The shaded region is the 95% confidence interval.

In addition to the FBI's hate crime data collection efforts, the Bureau of Justice Statistics (BJS) collects data on hate crime victimizations as part of the National Crime Victimization Survey (NCVS). NCVS respondents self-report their personal experiences with certain non-fatal crimes (e.g., robbery, aggravated assault) and, among responses to many other questions, whether they believe their criminal victimization was bias-motivated. In contrast to the FBI's data, NCVS victimizations include incidents that were reported and not reported to the police. This report presents analyses of hate crime data collected from 2015 to 2024 by both the FBI and NCVS.

Both the UCR and NCVS data indicate that hate crime rates have generally increased from 2015 to 2024. The UCR hate crime data show an 89% increase in the police-verified hate crime victimization rate over this 10-year period, rising from 1.8 hate crime victimizations per 100,000 people in 2015 to 3.4 per 100,000 in 2024. Similarly, the NCVS data show a 225% increase over this same 10-year period, rising from 55 self-reported hate crime victimizations per 100,000 people in 2015 to 179 per 100,000 in 2024. This trend of increasing hate crime rates observed in both datasets contrasts with current trends in non-bias-motivated crime rates, which have been generally decreasing in recent years.

The UCR and NCVS hate crime data show that most reported victimizations from 2015 to 2024 were motivated by race/ethnicity/ancestry. This was the bias category for 62.7% of victimizations reported in the UCR and 66.3% of victimizations reported in the NCVS. Other common bias categories were sexual orientation, religion, and gender identity. The UCR, but not the NCVS, reports more specific bias motivations within these broader categories. From 2015 to 2024, 51.0% of the 56,512 race/ethnicity/ancestry-motivated victimizations in the UCR data were anti-Black/African American, 56.9% of the 17,074 sexual-orientation-motivated offenses were anti-gay (male), 54.9% of the 13,098 religion-motivated offenses were anti-Jewish, and 78.3% of the 3,049 gender-identity-motivated offenses were anti-transgender.

---

Federal Hate Crime Data Collection

The Federal Bureau of Investigation (FBI) started its hate crime statistics program pursuant to the Hate Crime Statistics Act of 1990 (HCSA; P.L. 101-275). Currently, the act requires the U.S. Department of Justice (DOJ) to acquire and report data on crimes that "manifest evidence of prejudice based on race, gender and gender identity, religion, disability, sexual orientation, or ethnicity, including where appropriate the crimes of murder, non-negligent manslaughter; forcible rape; aggravated assault, simple assault, intimidation; arson; and destruction, damage or vandalism of property."¹ Since the enactment of the HCSA, the FBI has continually collected hate crime (alternatively referred to as bias-motivated crime) offense data submitted by state, local, federal, and tribal law enforcement agencies (LEAs) as part of its Uniform Crime Reporting (UCR) program. The FBI then compiles and publishes hate crime data annually, currently spanning the years 1991-2025.²

The Bureau of Justice Statistics (BJS) also collects data on hate crime victimizations as part of the National Crime Victimization Survey (NCVS). NCVS data are collected in partnership with the Census Bureau through semiannual interviews with residents of a nationally representative sample of households. Twice each year, the NCVS collects responses from approximately 240,000 individuals age 12 or older in 150,000 households.³ Respondents are asked about their personal experiences with certain non-fatal person crimes (i.e., rape or sexual assault, robbery, aggravated assault, simple assault, and personal larceny) and household property crimes (i.e., burglary/trespassing, motor vehicle theft, and household thefts) within the previous six months.

The NCVS also collects demographic information about individual victims (e.g., age, race, sex, disability status) and details about each crime incident, such as injuries to the victim, use of a weapon, and reporting to the police. This data collection effort intends to estimate the number of criminal victimizations in the United States each year, regardless of whether the victimizations were reported to police.

As part of the survey, the NCVS asks respondents who report victimizations if they believe their victimization was bias-motivated. If respondents report that they believe they were, they are then asked what evidence they have that the incidents were bias-motivated. Historically, BJS has identified an incident as a hate crime if the victim reported one of the following: (1) the police confirmed the victim was targeted for his or her characteristics or religious beliefs, (2) the offender(s) left behind hate symbols, or (3) the offenders(s) used hate language.⁴ For the purposes of this report, CRS used the same operational definition for identifying hate crimes as the NCVS.

Some policymakers have shown concern about the number and nature of hate crimes in the United States.⁵ This report explores data from the UCR and NCVS from 2015 to 2024 on the nature of hate crimes and describes the victims and perpetrators of these offenses.⁶ Though 2025 UCR hate crime data are available, as of the date of this report, the 2025 NCVS data have not been released. Thus, for the sake of equivalent comparison between the two data sources, CRS elected to use the 10-year range of 2015 to 2024 for the data in this report. The UCR and NCVS hate crime data were analyzed separately, but results (along with associated figures and tables) are presented in parallel to provide a more complete picture of hate crime in the United States. Following the review of the hate crime data, policy implications from the UCR and NCVS findings are briefly discussed.

Caveats and Limitations to Federal Hate Crime Data

UCR Hate Crime Data

There are important caveats and limitations to consider when interpreting the UCR hate crime data. First, these data represent only hate crimes reported to the police and subsequently reported to the FBI's UCR hate crime statistics program. UCR data do not account for criminal incidents not reported to the police, which may represent a substantial number of offenses. For example, BJS estimated that from 2015 to 2019 approximately 42% of hate crime victimizations were not reported to police.¹⁵

In addition to underreporting to police, the standards for vetting and investigating bias motivation in criminal incidents may affect the number of hate crimes reported to the UCR program each year. Specifically, the FBI lays out a two-tier decisionmaking process for investigating hate crimes. In the first tier, the responding officer must flag a criminal incident as a "suspected bias-motivated crime" if any bias indicators (e.g., defacing a building with a swastika) are present. In the second tier, a specially trained officer or unit must review the facts of the incident to make a final determination of whether a hate crime occurred.¹⁶

This comparatively higher standard of vetting and reporting (compared to non-bias-motivated crimes) may result in undercounting potential hate crimes reported to the FBI, particularly in smaller jurisdictions that lack funding, training, or staff to employ a unit of officers or even a single officer specially trained in hate crimes to complete the second tier of investigation. Furthermore, this higher standard of evidence may lead the UCR data to undercount hate crimes against groups that are not associated with obvious bias indicators. For example, the swastika is a well-known bias indicator for anti-Jewish sentiment, but few equally well-known symbols represent other religion-based bias motivations (e.g., anti-Buddhist, anti-Protestant). Thus, the two-tier decisionmaking process may systematically undercount certain bias-motivated crimes against certain populations with less obvious bias indicators.

Another caveat is that these data represent submissions only from the LEAs that participate in the FBI's UCR program and submit hate crime data. Law enforcement participation in the FBI's UCR program is voluntary, and not all LEAs participate. However, with the exception of a few years, there has been a general trend of increasing numbers of LEAs participating in the UCR hate crime statistics program each year over the past decade.¹⁷ In its UCR Summary of Reported Crimes in the Nation, 2024, the FBI reported that 16,419 LEAs—or 84.9% of all LEAs enrolled in the UCR program—participated in the hate crime statistics program in 2024. This amounted to a 2.6% increase from the previous year's report, and, taken together, all participating agencies cover 95.1% of the total U.S. population.¹⁸ However, as some researchers have noted, jurisdiction-specific hate crime reporting policies vary.¹⁹ Some states (e.g., New Jersey) have policies that explicitly require their LEAs to investigate hate crimes according to specific standards and report these incidents to the state agencies responsible for submitting these data to the FBI.²⁰

When an agency reports that no known hate crimes occurred in its jurisdiction for a given year, this is known as a zero report. Of the 16,419 LEAs that participated in the UCR program in 2024, 81% submitted zero reports. This is not anomalous, as most LEAs participating in the UCR hate crime statistics program in any given year submit zero reports as opposed to reporting data on hate crimes. Thus, even though participation among LEAs has generally increased in recent years, a relatively small percentage of them (about 2,127 per year on average) contribute data pertaining to reported hate crime incidents each year (see Figure A-1 in Appendix A for a visual depiction of hate crime reporting from 1996 to 2024).²¹

Additionally, since 2015, the percentage of agencies reporting at least one hate crime incident has varied considerably depending on the type of agency (e.g., city, county, state). The percentage of state police agencies reporting one or more hate crimes was higher than any other LEA type from 2015 to 2024. However, the relatively high annual percentage of state LEAs reporting at least one hate crime is partially due to the relatively low number of state police agencies compared to most other types of LEAs (see Figure A-2 in Appendix A for a visual depiction of hate crime reporting by agency type from 2015 to 2024).

There are also limitations on the information available from the FBI's publicly available hate crime data. For instance, though some offender demographics (e.g., race) are included, little to no information about victims is included. Additionally, no arrest information is included, meaning the data do not indicate whether any offenders were arrested and charged for the reported hate crime incidents. Thus, the UCR hate crime data primarily provide information about the characteristics of the criminal incident (e.g., bias motivation, number and type of offense, number of victims) rather than the persons involved.

NCVS Hate Crime Data

There are also important caveats and limitations to consider when interpreting the NCVS hate crime data. First, individuals self-report their criminal victimizations through structured interviews with Census Bureau representatives. Though interviewing crime victims helps account for incidents that were not reported to the police, there are data quality concerns associated with this approach, such as victims potentially misremembering relevant details about criminal incidents. BJS tries to limit recall bias by (1) having interviewers clearly explain to the respondent what a bias-motivated crime is and (2) conducting interviews at six-month intervals (rather than annually). In the case of bias-motivated crimes specifically, there is also the issue that, unlike the UCR data, the NCVS does not employ the two-tier decisionmaking process that public safety officials use to determine whether a criminal incident was bias-motivated.

The NCVS interviews only individuals age 12 and older living in non-institutional settings.²² It therefore does not capture crimes committed against young children, people experiencing homelessness, or people living in correctional facilities, nursing homes, military bases, and other institutional settings.²³ The NCVS also covers fewer types of crimes than UCR data does. Additionally, because these data are collected from a sample, statistical estimates produced from NCVS data have a margin of error that is important to consider when comparing estimates of specific variables.²⁴ These data cannot always produce reliable estimates for subpopulations with small sample sizes (e.g., people with disabilities).

In 2024, BJS redesigned the NCVS in an effort to improve the efficiency, reliability, and utility of the survey.²⁵ The survey was administered using a split sample design, where approximately half of the surveyed households were interviewed using the legacy survey instrument (the questions that were asked in previous years) and the other half was surveyed using the redesigned instrument. The split was designed so that each half sample would be similar to the samples from recent years, and BJS included a set of survey weights that adjusts the half samples to U.S. population totals. In this report, CRS used the legacy instrument to avoid concerns about whether the data was comparable to previous years. However, the 2024 half sample was smaller (109,341 persons age 12 or older in 69,396 households) than the sample from previous years (approximately 240,000 persons in approximately 150,000 households), so the standard errors for 2024 are larger.

See Table 1 for a direct comparison of elements found in the UCR and NCVS data.

Findings from UCR and NCVS Hate Crime Data

UCR and NCVS Victimizations, 2015-2024

Over the 2015-2024 period, law enforcement reported 90,143 verified hate crime victimizations to the FBI's UCR hate crime statistics program. By contrast, over the same period, the NCVS data yielded an estimate of 2,525,312 victimizations that BJS identified as hate crimes, a number over 28 times larger than what is indicated by UCR data. Figure 1 presents NCVS and UCR hate crime victimization data. For the NCVS, it presents the number of self-reported hate crime victimizations, the number of those victimizations BJS identified as hate crimes, and the number that were reported to the police. For the UCR, it presents the number of reported hate crimes that were submitted to the FBI and the number that met the FBI's UCR program criteria for assessing whether the crime was bias-motivated.

Figure 1. Hate Crime Victimizations, 2015-2024 Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: The 464,520 hate crime victimizations reported to UCR represent the 87,970 victimizations verified by the UCR's two-tier process plus 376,550 victimizations that were marked as "unknown" bias motivation in the 2015-2024 data. As per UCR guidance, law enforcement agencies use the "unknown" entry under bias motivation when facts regarding bias in a criminal incident are ambiguous and require further investigation. Given the indirect manner from which this estimate of 464,520 hate crime victimizations reported to UCR had to be inferred, it should be interpreted as an estimate as opposed to a known value. Additionally, there is no way to track specific incidents or victimizations between the UCR and NCVS data, thus CRS could not determine which (if any) of the hate crime victimizations reported to the NCVS were then also reported by law enforcement to the UCR program. This figure represents a conceptual relationship between the two datasets as opposed to a literal one-to-one mapping of individual incidents.

Overall Trends from 2015 to 2024

According to UCR hate crime data from 2015 to 2024, there were 91,559 hate crime incidents reported to the police, with 95,222 associated offenses affecting a total of 90,143 individual victims.²⁶ Figure 2 shows that hate crime incident rates (per 100,000 people) rose over this 10-year period, increasing from 1.8 incidents per 100,000 in 2015 to a peak of 3.5 per 100,000 in 2023. However, the rate declined from 3.6 incidents per 100,000 people in 2023 to 3.4 per 100,000 in 2024, amounting to a 3% decrease in number of incidents. Academic researchers²⁷ and advocacy groups²⁸ have also noted the increasing rate of hate crime incidents over this period.

UCR victimization and offense rates have largely followed the trend in incident rates, with offense rates generally being slightly higher than incident or victimization rates (Figure 2). The similarity among these rates is important to keep in mind: Though victimizations will be the primary unit of analysis from the UCR data throughout the rest of this report (except where indicated otherwise), patterns from the UCR hate crime data remain largely consistent whether looking at incidents, victimizations, or offenses (see the Key Definitions text box in the "Federal Hate Crime Data Collection" section for further explanation of these units of analysis).

Further analysis revealed that this trend of increasing hate crime incident rates was statistically significant, even when accounting for overall crime rates (violent and property crime),²⁹ with the hate crime incident rate increasing by 0.26 incidents per 100,000 people per year on average (see Table B-2). Parallel analyses confirmed both the victimization and offense rate increases to also be statistically significant, even when controlling for overall crime rates. In other words, hate crime incident, victimization, and offense rates all increased from 2015 to 2024 independently from overall crime rates across this period (see Table B-2).

More recently, however, the 2025 UCR data—which is not covered in the analyses in this report for reasons discussed in "Federal Hate Crime Data Collection"—show that hate crime incidents have decreased by 7.0% from 2024 to 2025. Offenses and victimizations similarly decreased from 2024 to 2025.³⁰ Thus, the trend of increasing hate crime rates described herein may be limited to the period of time (i.e., 2015 to 2024) analyzed in this report and should thus be interpreted with this potential caveat in mind.

Of the 91,559 total UCR hate crime incidents reported to the police from 2015 to 2024, individuals were by far the most common victim type (79.6% of incidents), with businesses being the next most common (7.4%). Most of the UCR hate crime incidents during this period (68.6%) involved single individual victims.

Figure 2. Hate Crime Incident, Victimization, and Offense Rates (per 100,000 People), 2015-2024 Uniform Crime Reporting Program

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025).

Analysis of the NCVS data from 2015 to 2024 revealed higher rates of hate crime victimization than what UCR data indicated. Figure 3 shows the hate crime victimization rate from 2015 to 2024.³¹ This rate steadily increased from 2015 to 2019, decreased from 2019 to 2020, remained relatively constant until 2023, and then spiked in 2024 to 179 victimizations per 100,000 people.

Figure 3. Hate Crime Victimization Rates, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The shaded region is the 95% confidence interval.

Incidents involving hate language as evidence of bias motivation largely drove the increase in NCVS hate crimes against individuals from 2023 to 2024. Figure 4 shows the rates of hate crime victimizations reported in the NCVS from 2015 to 2024, broken down by the type of evidence the victim reported.³² The use of hate language was the most common type of evidence, as 99% of hate crimes reported during the 10-year period involved the use of hate language as one of the pieces of evidence, including 100% of the hate crimes reported in 2024.

Figure 4. Hate Crime Victimization Rate (per 100,000 People), by Type of Evidence, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The shaded region is the 95% confidence interval. An asterisk (*) indicates that the estimate is unreliable due to having a relative standard error that exceeds 50.

As mentioned in the "NCVS Hate Crime Data" section above, an important feature of the NCVS data is that they capture crime victimizations that may not have been reported to police. Table 2 displays the number of NCVS hate crime victimizations reported and not reported to the police during 2015-2024 and, for those who did not report their victimization to the police, the most important reason why. A little over two in five hate crime victimizations (47.1%) were reported to the police. For comparison, BJS reported that, from 2015 to 2019, 57.3% of violent hate crime victimizations were reported to police.³³ Exploratory analysis of the 2015-2024 NCVS data conducted by CRS revealed that the percentage of hate crime victimizations reported to police varied considerably year to year relative to the overall percentage of crime victimizations reported to police (see Figure A-3). Approximately 387,000 (29.6%) unreported hate crime victimizations were not reported to the police because they were reported to other officials (e.g., a school official), making this the most common reason for not reporting hate crimes to law enforcement (see Table 2).

Victim Characteristics

Victim Age

From 2015 to 2024, 9,909 (11.0%) of UCR hate crime victimizations involved juvenile victims.³⁴ Children ages 12-17 experienced the highest rate of hate crime victimizations (168 per 100,000) reported to the 2015-2024 NCVS. Hate crime victimization rates for adults ages 30-64 and children ages 12-17 in the NCVS were relatively stable for four years between 2020 and 2023, then spiked in 2024 (Figure 5).

Figure 5. Hate Crime Victimizations by Victim's Age, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval. An asterisk (*) indicates that the estimate is unreliable due to having a relative standard error that exceeds 50.

Victim Sex³⁵

The NCVS data show there were changes in the sex distribution of hate crime victims from 2015 to 2024. Figure 6 shows the number of hate crime victimizations by the sex of the victim in 2015 and 2024. In 2015, approximately two in every three hate crime victims (67.0%) were male. Over the course of the decade, this pattern reversed so that nearly two of every three hate crime victims (63.2%) in 2024 were female. In addition, Figure 7 shows that there was a six-fold increase in the female hate crime victimization rate from 2015 (35.3 per 100,000) to 2024 (221.6 per 100,000), compared to a 76% increase in the male hate crime victimization rate (75.5 per 100,000 in 2015 to 134.1 per 100,000 in 2024). This 2024 pattern of hate crime victimization differs from broader violent crime victimization trends in the NCVS data, which showed a slightly higher violent victimization rate for men (approximately 2,470 victimizations per 100,000 people) compared to women (approximately 2,190 victimizations per 100,000 people) in 2024.³⁶

Figure 6. Hate Crime Victimizations by Victim's Sex, 2015 and 2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Figure 7. Hate Crime Victimizations by Victim's Sex, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The shaded region is the 95% confidence interval.

Victim Race and Ethnicity³⁷

Figure 8 shows NCVS hate crime victimization estimates for people in different racial and ethnic groups. For the purposes of this analysis, individual victims were divided into six mutually exclusive categories: Hispanic (any race), non-Hispanic White, non-Hispanic Black, non-Hispanic Asian, non-Hispanic American Indian or Alaska Native, and non-Hispanic Mixed Race.

Figure 8 shows both the number and rate (per 100,000 people) of hate crime victimizations by race/ethnic group. By raw counts, non-Hispanic Whites had the highest number of hate crime victimizations during 2015-2024, whereas non-Hispanic American Indians or Alaska Natives had the lowest number. However, when accounting for the population of the respective racial/ethnic group, victimization rates by racial/ethnic group were highest for non-Hispanic American Indians or Alaska Natives and lowest for non-Hispanic Asians. Non-Hispanic Whites had the second-lowest rate.

Figure 8. Hate Crime Victimizations by Victim's Race/Ethnicity, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Incident Characteristics

From 2015 to 2024, most bias-motivated offenses reported to the UCR program involved crimes against persons (58,344 offenses, or 61.5% of all offenses), followed by crimes against property (34,432 offenses, or 36.3% of all offenses) and crimes against society³⁸ (2,125 offenses, or 2.2% of all offenses). Most bias-motivated crimes against persons were assault offenses,³⁹ whereas destruction of property (which includes vandalism and damage of property) was the most common bias-motivated offense against property. As shown in Figure 9, bias-motivated assault and destruction of property offense rates (i.e., number of offenses per 100,000 people) gradually increased from 2015 to 2024, whereas other offense types (e.g., larceny/theft) generally remained consistent.

Figure 9. Bias-Motivated Offense Rate by Offense Type, 2015-2024 Uniform Crime Reporting Program

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025). Notes: "Other Offenses" is an ad hoc category that includes offense types such as fraud and sex offenses, each of which made up fewer than 1% of all recorded offenses on their own. These were grouped together into a single category to increase legibility of the figure.

Although the absolute number of offenses differs between UCR and NCVS datasets, both sources generally agree on the most common types of offenses. Figure 10 shows the number of hate crime victimizations reported in the 2015-2024 UCR and NCVS data by type of offenses committed. Intimidation was the most frequent type of offense in the UCR data, whereas simple assault was the most frequent in the NCVS data⁴⁰—specifically, 37.8% of UCR hate crime victimizations were intimidation, and 67.0% of NCVS hate crime victimizations were simple assaults.⁴¹ Some apparent differences between the offense types in the UCR and NCVS data emerged. For instance, bias-motived robbery was relatively more common in the NCVS data (10.7% of bias-motivated victimizations) compared to the UCR data (2.0% of reported hate crime victimizations) in 2015-2024.

Figure 10. Hate Crime Victimizations by Offense Type, 2015-2024 Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Due to a combination of underreporting victimizations to the police and the FBI's two-tier process for verifying hate crimes, there are substantially fewer hate crime victimizations reported in the UCR data than in the NCVS data. However, there are more offense types in UCR than NCVS. For instance, there is no analogue to "Intimidation" or "Destruction/Damage/Vandalism of Property" among the subset of NCVS data analyzed in this report. The "Other" category includes the following UCR offenses: Prostitution Offenses, Animal Cruelty, Human Trafficking, Bribery, Pornography/Obscene Material, Embezzlement, Extortion/Blackmail, Drug/Narcotic Offenses, Counterfeiting/Forgery, and None/Unknown. NCVS estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Figure 11 reports hate crime victimizations by the victim's sex and offense type in the 2015-2024 NCVS. The sex distribution was nearly even for simple assault, with male victims accounting for 48.9% of simple assault hate crime victimizations. For aggravated assaults, most victimizations had male victims (56.0%). Most rape or sexual assault hate crime victimizations (62.8%) and most robbery hate crime victimizations (59.1%) had female victims.

Figure 11. Hate Crime Victimizations by Victim's Sex and Offense Type, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Figure 12 shows that the most common bias category across both data sources was race/ethnicity/ancestry. The victim's perceived race and ethnicity were each reported as a motivating bias in approximately 1.7 million hate crime victimizations from 2015 to 2024 in the NCVS and 57,282 reported victimizations in the UCR data. The NCVS data indicated that race or ethnicity was a motivating bias in 66.3% of victimizations. By comparison, UCR data indicated race/ethnicity/ancestry was a motivating bias in 63.5% of victimizations.

The next most common motivating bias according to the 2015-2024 NCVS data was the victim's association with people who have certain characteristics or religious beliefs (26.5%), for which there is no equivalent bias category in the UCR data (Figure 12). Victims' gender was the third most common bias category in the NCVS, being listed as a bias motivation in over one-fifth of all hate crime victimizations. By comparison, gender was a relatively rare bias category in the UCR data, at 1.1% of victimizations. The fourth most common bias category in the NCVS data was the victim's sexual orientation, at 17.5% of 2015-2024 victimizations. In contrast, sexual orientation was the second most common bias category in the UCR at 19.1% of 2015-2024 victimizations.

Anti-Black or African American victimizations were the most common in the race/ethnicity/ancestry bias category, at 32.0% of all hate crime victimizations. Anti-gay (male) victimizations were the most common in the sexual orientation category, at 10.8% of all victimizations. Anti-Jewish were the most common among the religion category, at 8.0% of victimizations. Anti-transgender were the most common among the gender identity category, at 2.6% of victimizations. Anti-mental disability were the most common among the disability category, at 1.0% of victimizations. Anti-female were the most common among the gender category, at 0.9% of victimizations. See Table A-1 in Appendix A for a breakdown of all victimizations by bias category and motivation in the UCR data over the entire 10-year period from 2015 to 2024 and Table A-2 for all victimizations by bias category in the NCVS data for 2024.

Figure 12. Hate Crime Victimizations by Bias Category, 2015-2024 Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Due to a combination of underreporting victimizations to the police and the FBI's two-tier process for verifying hate crimes, there are substantially fewer hate crime victimizations reported in the UCR data than in the NCVS data. NCVS estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Findings from UCR and NCVS data differed regarding where most hate crime victimizations took place. Figure 13 shows the number of hate crime victimizations reported in the 2015-2024 UCR and NCVS data by types of locations. The most common location for hate crimes reported in both datasets was private residences. In the NCVS the second most common location was roadways (20.3%). According to the UCR data, roadways (19.2%) and parking lots (6.2%) were common locations as well.⁴²

Figure 13. Hate Crime Victimizations by Type of Location, 2015-2024 Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Due to a combination of underreporting victimizations to the police and the FBI's two-tier process for verifying hate crimes, there are substantially fewer hate crime victimizations reported in the UCR data than in the NCVS. NCVS estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence interval.

Offender Characteristics

Limiting the analysis of 2015-2024 UCR and NCVS hate crime data to incidents with single offenders (as opposed to multiple offenders),⁴³ Figure 14 compares the race/ethnicity of offenders in the two datasets for that year. Though the raw numbers differ, the patterns are nearly identical between the UCR and NCVS, with non-Hispanic Whites making up most hate crime offenders.

There is a particular caveat to bear in mind when interpreting UCR offender-level data in the current analyses. When offender information is reported to the police (referred to in the data as a "known offender"), that does not mean that the offender or offenders have been positively identified. Rather, law enforcement gather this information from victims, witnesses, and other sources at the time of the offense. Thus, most offender-level information, such as race, is presumptive in nature and based on information available to officers at that time. As the FBI explains:

In the UCR Program, the term known offender does not imply the suspect's identity is known; rather, the term indicates some aspect of the suspect was identified, thus distinguishing the suspect from an unknown offender. Law enforcement agencies specify the number of offenders and, when possible, the race, ethnicity, and age of the offender or offenders as a group.⁴⁴

The NCVS contains questions about offenders involved in criminal incidents. Respondents are first asked the following question: "Do you know or have you learned anything about the offender(s)—for instance, whether there was one or more than one offender involved, whether it was someone young or old, or male or female?"⁴⁵ If respondents answer yes, they are asked a series of funneled follow-up questions, such as how they learned about the offender (e.g., firsthand seeing or hearing offender, learned from another eyewitness, from police), whether they knew the offender, and basic demographic information (if known). Thus, like the UCR offender information, these NCVS offender characteristics, such as sex or race/ethnicity, reflect the respondents' own knowledge and memories of the facts.

Figure 14. Reported Race of Hate Crime Offenders, Single Offender Incidents Only, 2015-2024 Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: This figure excludes data from incidents that were recorded as having more than one offender. The UCR known offender data only reports one race and one ethnicity per incident, even when there are multiple offenders. The NCVS data reports the number of single-offender hate crime incidents where the offender is listed as having a particular race/ethnicity. CRS was unable to calculate standard errors for this analysis because BJS provides only the information needed to calculate standard errors for victimizations, not incidents. Accordingly, these estimates should be interpreted with caution.

Comparing single and multiple offender victimizations in the 2015-2024 NCVS data, Figure 15 shows that over half of all hate crimes were reported to have lone male offenders. The second most common reported hate crime offender type during 2015-2024 was a single female offender (15.7%).

Figure 15. Hate Crime Victimizations by Reported Offender's Sex, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older. The error bars (black lines) report the 95% confidence intervals.

Proportionality of Hate Crime Victimizations by Race/Ethnicity and Religion Compared to the U.S. Population

Looking at the number of victimizations by bias motivation from both the UCR and NCVS data (see Figure 12) might raise questions for policymakers about the degree to which some groups are disproportionately targeted by bias-motived crimes compared to their relative shares of the population.

Using five-year estimates from the Census Bureau's 2024 American Community Survey (ACS),⁴⁶ CRS compared race/ethnicity/ancestry-motivated hate crimes targeting individuals⁴⁷ in the 2020-2024 UCR and NCVS data to ACS population demographics for the corresponding groups.⁴⁸ More specifically, UCR race/ethnicity/ancestry victimizations were compared to total ACS population estimates broken down by race, whereas NCVS victimizations were compared to the ACS population estimates limited to those 12 years and older in order to more accurately compare to the NCVS sample parameters. Figure 16 shows two key patterns when comparing 2020-2024 UCR race/ethnicity/ancestry victimizations, 2020-2024 NCVS race and ethnicity victimizations, and the corresponding ACS population estimates.

First, of all hate crime victimizations motivated by race/ethnicity, the percentage of anti-White victimizations from 2020 to 2024 as measured by both the UCR and NCVS data were disproportionally low compared to the average percentage of non-Hispanic Whites in the population of the United States during the period. However, the pattern is more complex looking at anti-Black and anti-Hispanic victimizations. Specifically, the NCVS data showed that of all hate crime victimizations motivated by race/ethnicity, the percentage of anti-Black victimizations was roughly proportional to the Black share of the U.S. population age 12 and older, whereas the UCR data showed a disproportionately high percentage of anti-Black victimizations compared to their share of the U.S. population. An opposite pattern emerged between the two data sources concerning anti-Hispanic victimizations, with the UCR data indicating that the percentage of anti-Hispanic victimizations was roughly proportional to the Hispanic share of the U.S. population, whereas the NCVS data showed a disproportionately high percentage of anti-Hispanic victimizations compared to the Hispanic share of the U.S. population age 12 and older. A statistical analysis confirmed that both the UCR and NCVS racially motivated victimizations overall are disproportionate to corresponding Census estimates (see Table B-1).

Second, the number of 2020-2024 UCR and NCVS race/ethnicity/ancestry victimizations were disparate from each other in some cases. Specifically, proportionally more anti-Black bias-motivated victimizations were reported to the UCR program than to the NCVS, whereas proportionally more anti-White and anti-Hispanic bias-motivated victimizations were reported to the NCVS than to the UCR. One important caveat is that for the NCVS data, specific bias motivations were inferred based on the respondent's own self-reported race.

Figure 16. 2020-2024 U.S. Population Racial Demographic Percentages Compared to 2020-2024 Race/Ethnicity/Ancestry Hate Crime Victimization Percentages Uniform Crime Reporting (UCR) Program and National Crime Victimization Survey (NCVS)

Source: FBI, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); National Crime Victimization Survey, https://www.icpsr.umich.edu/web/NACJD/studies/39273; and 2024 American Community Survey (ACS) five-year estimates from Kyle Walker et al., tidycensus: Load US Census Boundary and Attribute Data as 'tidyverse' and 'sf'-Ready Data Frames, R package version 1.7.5, 2026, https://walker-data.com/tidycensus/. Notes: All ACS racial categories (aside from Hispanic) are the single-race, non-Hispanic estimates for that group for the purposes of creating directly comparable, mutually exclusive race categories that approximated the UCR and NCVS conceptualizations of race/ethnicity. Due to low numbers of Native Hawaiian/Pacific Islander individuals in the 2020-2024 NCVS sample, NCVS estimates for this group are not included.

The Census Bureau does not collect data on religious affiliation. However, other nongovernment entities conduct surveys on this topic that can provide an estimate of population-level religious affiliation in the United States. One commonly cited religious affiliation survey is Pew Research Center's Religious Landscape Study, which surveyed approximately 36,000 respondents across the United States in 2023 and 2024 and asked various questions about religious affiliation and practice.⁴⁹ These religious affiliation estimates from Pew and the corresponding percentages of religion-specific hate crimes targeting individuals⁵⁰ in the 2023 and 2024 UCR data were compared.⁵¹ As shown in Figure 17, the percentage of all hate crime victimizations motivated by anti-Protestant, Catholic, and Atheist/Agnostic bias (1.4%, 1.4%, and 0.8% of all religion-motivated hate crime victimizations, respectively) was disproportionately low compared to the estimated percentage of the population claiming Protestant, Catholic, and Atheistic/Agnostic affiliation (40.0%, 18.9%, and 11.0% of the weighted survey respondents, respectively). On the other hand, the percentage of all hate crime victimizations motivated by anti-Jewish and Muslim bias (70.8% and 14.5% of all religion-motivated hate victimizations, respectively) was disproportionately high compared to the surveyed Jewish and Muslim populations of the United States (1.7% and 1.2% of the weighted survey respondents, respectively). All other religion bias motivations were approximately proportional to those religions' estimated share of the population in 2023 and 2024. A statistical analysis conducted by CRS also indicated that the UCR religion-motivated victimizations were overall disproportionate to the corresponding Pew estimates (see Table B-1).

There are many important caveats and limitations to the data presented in this section. Due to the nature of Census and Pew data collection, these data represent very specific demographic moments. U.S. racial demographics and religious affiliation trends are dynamic and may continue changing in the future, limiting the future generalizability of these patterns. Thus, these findings should be interpreted as a snapshot of these points in time rather than stable trends over time.

Figure 17. 2023-2024 Pew Religious Affiliation Estimates Compared to 2023-2024 Religion Hate Crime Victimization Percentages Uniform Crime Reporting Program

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed August 29, 2025); and Pew Research Center, "2023-24 Religious Landscape Study (RLS) Dataset," 2025, doi: 10.58094/3kwb-bf52.

There are some possible explanations for the patterns of hate crime victimization discussed in this section and throughout the rest of this report. The first is that broader social and environmental factors may be at least partially influencing the hate crime victimization patterns presented in this report. Other researchers have observed that certain racial, ethnic, and religious groups are disproportionately targeted for hate crimes depending on societal trends.⁵² For instance, some research has found that as minority groups increase as a proportion of the population in an area, so too does their risk of hate crime victimization.⁵³ Furthermore, influential factors may differ based on the specific bias motivation. In its guide on hate crimes, the Center for Problem-Oriented Policing offers the following analysis of the issue:

Anti-black hate crime usually occurs in relatively organized communities with high levels of informal social control. In contrast, anti-white hate crimes are more likely to occur in disorganized locations where residential turnover is more common. Meanwhile, anti-gay hate crimes are more likely to occur in areas where gays are more numerous. Again, context is important. It has been argued that local hate crimes against Jews are linked to a rise in tension or a specific event in the Israel-Palestine conflict or other tensions occurring in the Middle East. Similarly, some find a correlation between the demonization of Muslims in the media and the victimization of local Muslims by hate crimes.⁵⁴

A second explanation to consider is that these patterns could be at least partially reflective of group differences in reporting hate crimes. For instance, past research has found that Black individuals, particularly Black women, are more likely than persons in other demographic groups to report criminal victimizations to the police.⁵⁵ Exploratory analyses of victim race/ethnicity in the 2020-2024 NCVS data also suggested that Black victims were the most likely of any racial group to report their hate crime victimizations to police (see Table A-4). This possibility, some could argue, may partially explain why anti-Black victimizations comprised the most common bias motivation in the 2015-2024 UCR data.

These trends could also reflect the inherently empirical issues related to identifying and reporting hate crime victimizations. For example, in its Hate Crime Data Collection Guidelines and Training Manual, the FBI states that "bias-related oral comments" such as racial slurs and epithets, among other aspects of the crime, can be objective evidence of a crime being motivated by bias.⁵⁶ Certain racial or religious groups have a wider lexicon of slurs and hate symbols directed against them, meaning it is easier to identify that a criminal act was bias-motivated against that group. For instance, if someone wearing a shirt depicting a swastika assaults a Jewish person leaving a synagogue, it may be much easier for law enforcement and the victim to determine that it was an anti-Jewish-motivated assault. In comparison, if someone assaults a Buddhist leaving a temple, it might be more difficult to determine whether it was a bias-motivated crime given that there is not a symbolic equivalent to the swastika for anti-Buddhist sentiment that is a widely known one used in a Western context.

Policy Considerations for Congress

As shown in the analyses throughout this report, increasingly more hate crime incidents have been reported annually to the UCR hate crime statistics program and the NCVS over the course of the past 10 years (see Figure 2 and Figure 3). This trend coincides with a time when rates of reported non-bias-motivated property and violent crimes have been decreasing.⁵⁷ The number of agencies participating in the UCR program increased from 2015 to 2024 as well (see Figure A-1 and Figure A-2). Thus, it is plausible that the increase in reported hate crimes over the past 10 years in the UCR data could be partially due to improved LEA reporting during this time. However, the greater number of LEAs reporting hate crime data to the UCR program does not account for the increase in self-reported hate crime victimization found in the NCVS data during this same period.

As shown in the present findings, bias-motivated crimes seem to be distinct from non-bias-motivated crime, and, at times, the two types of crime show disparate patterns. CRS analyses indicated that overall crime rates were not statistically significantly associated with UCR bias-motivated crime rates from 2015 to 2024 (see Table B-2). The analysis of the data also shows divergent trends among hate crime as compared to other types of crime, particularly in more recent years. For instance, analysis of the 2024 NCVS data showed different victim sex ratios for bias-motivated offenses compared to what is typically observed in similar non-bias-motived crimes. Additionally, the fact that juveniles and women made up such a sizeable portion of the NCVS victim pool suddenly in 2024 (see Figure 5, Figure 6, and Figure 7) may be noteworthy. These findings might indicate something about juvenile delinquency, school bullying, and violence against women intersecting with increasing online radicalization, extremism, and hate.⁵⁸ The data also seem to indicate that certain groups (e.g., people of Jewish or Muslim faith) are disproportionately targeted in hate crimes, particularly considering their relative share of the U.S. population (see Figure 16 and Figure 17).

There are several issues policymakers might consider if they choose to address issues around hate crimes and hate crime reporting. First, Congress could address the potential issues with how LEAs report hate crime to the FBI. Recent legislative efforts to incentivize better reporting among local LEAs include, for example, the Improving Reporting to Prevent Hate Act of 2025 (H.R. 2588, S. 3724), which would require DOJ to proactively audit hate crime data submissions from LEAs serving jurisdictions of more than 100,000 residents for accuracy. Under the proposal, any LEAs found not to be "credibly" reporting hate crimes would risk losing their Edward Byrne Memorial Justice Assistance Grant funding.⁵⁹

Congress could also develop legislation that would specify which types of bias motivations are included in hate crime data collected by the FBI and/or BJS. Over the past 25 years the FBI has added new bias categories and types (e.g., adding the anti-Arab bias motivation in 2015) to its hate crime reporting program. Further changes may come considering the Office of Management and Budget (OMB) issued a new Statistical Policy Directive that updates federal standards for collecting and presenting data on race and ethnicity. Of note, OMB added two new racial categories for federal statistical reporting: Hispanic or Latino (which was previously categorized as an ethnicity) and Middle Eastern or North African (which includes groups that had previously been grouped under White).⁶⁰ Still, Congress could pass legislation requiring either agency to collect data on additional bias motivations. For instance, under some state hate crime laws, political expression (e.g., Utah) and immigration status (e.g., Illinois) are considered bias categories.⁶¹

Congress could also consider legislation to address hate crimes themselves more directly. For example, given the findings of the disparity analysis presented in the "Proportionality of Hate Crime Victimizations by Race/Ethnicity and Religion Compared to the U.S. Population" section, policymakers may consider targeting resources and programs for groups disproportionally at risk of hate crime victimization. An example of an existing grant program that has promoted similar policy aims by focusing resources on certain at-risk populations is the Nonprofit Security Grant Program. According to the Federal Emergency Management Agency, this program "provides funding support for target hardening and other physical security enhancements and activities to nonprofit organizations that are at high risk of terrorist attack."⁶²

Funds could also be designated toward grant programs aimed at training police officers in investigating and preventing hate crimes. This could be especially helpful for smaller LEAs that lack training on how to investigate bias motivation. Up-to-date training could also keep officers vigilant of the high-risk targets of hate crime (e.g., increased anti-Jewish or anti-Muslim hate crimes in response to conflict in the Middle East) and ever-evolving indicators of bias motivation (e.g., new hate symbols or slurs circulated on the internet). Some of these funds could also be directed toward community outreach, encouraging people to report hate crimes to police. Exploratory analyses of the 2015-2024 NCVS data showed that, compared to overall crime victimizations, the annual average rate of reporting of hate crime victimizations to police was more inconsistent and seems to be trending downward in recent years (see Figure A-3). Though this trend is difficult to interpret without further data, it may be an indicator of fluctuating trust some communities have in the ability of law enforcement to effectively handle hate crime investigations. Proactive law enforcement outreach on the issue of hate crime may encourage more consistent reporting from victims.

Congress could also provide funding for grants programs to prevent and deter crimes against potentially vulnerable populations. The Matthew Shepard and James Byrd, Jr. Hate Crimes Program is one that has been directed at this purpose.⁶³ Given the finding from the NCVS data that schools are a common location for hate crime victimizations (see Figure 13), Congress could also consider legislation to provide funding to public schools for hate crime awareness, de-radicalization training, and increased security measures. Programs could focus on training teachers and administrators on which groups of students may be at highest risk and how to watch for signs of hate ideology on school grounds and among their students, such as internet-sourced racial slurs.

Congress might also decide to defer to state-level policymakers to address hate crimes in their respective states. Like most other crimes, hate crimes by and large are investigated and prosecuted by state and local authorities. While some of the findings presented in this report indicate a recent increase in hate crime incidents, bias-motivated crimes are still reported to the UCR and NCVS less frequently compared to non-bias-motivated crimes. Thus, Congress could decide to leave further legislation on this topic up to the individual states.

Appendix A. Hate Crime Figures and Tables

Supplemental Figures

Figure A-1. Number of Law Enforcement Agencies by Participation Status in the Hate Crime Statistics Program, 1996-2024 Uniform Crime Reporting Program

Source: U.S. Department of Justice, Federal Bureau of Investigation (FBI), Hate Crime Statistics and UCR Program Participation Data for 1996-2024, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads. Notes: These data represent the sum of federal, state, local, and tribal law enforcement agencies across the United States and its territories. Note that some of the non-participating agencies captured in these yearly totals may not be agencies where it would be reasonably expected for them to handle hate crime cases as part of their regular functions (e.g., the Internal Revenue Service Criminal Investigation Unit, special jurisdiction agencies such as airport police). The FBI accepted only National Incident-Based Reporting System data and no Summary Reporting System data in 2021 before reverting to accepting both formats from 2022 onward. This likely contributed to a noticeable reduction in hate crime data submissions in 2021.

Figure A-2. Percentage of Agencies Reporting ≥ 1 Hate Crime by Agency Type, 2015-2024 Uniform Crime Reporting Program

Source: Federal Bureau of Investigation, Crime Data Explorer, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads (accessed May 13, 2025, and August 29, 2025).

Figure A-3. Hate Crime Victimizations Versus All Personal Criminal Victimizations Reported to Police, 2015-2024 National Crime Victimization Survey

Source: CRS, using data from the Bureau of Justice Statistics, National Crime Victimization Survey, https://bjs.ojp.gov/data-collection/ncvs. Notes: Estimates are based on the non-institutionalized U.S. resident population age 12 or older.

Descriptive Statistics

Appendix B. Inferential Statistics

Footnotes

1.	34 U.S.C. §41305.
2.	For more information on how and why the FBI collects hate crime data, see CRS Report R46318, Federal Data on Hate Crimes in the United States, by Nathan Kemper.
3.	BJS, "National Crime Victimization Survey (NCVS)," 2024, https://bjs.ojp.gov/data-collection/ncvs.
4.	Grace Kena and Alexandra Thompson, Hate Crime Victimization, 2005-2019, BJS, September 2021, p. 15, https://bjs.ojp.gov/library/publications/hate-crime-victimization-2005-2019.
5.	For example, see the Improving Reporting to Prevent Hate Act of 2026 (S. 3724, 119th Congress), a bill aimed at incentivizing more complete hate crime reporting from law enforcement.
6.	FBI hate crime data are available at https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads, and NCVS data are available at https://www.icpsr.umich.edu/web/NACJD/studies/39273.
7.	FBI, UCR Summary of Reported Crimes in the Nation, 2024, August 2025, p. 37, https://cde.ucr.cjis.gov/LATEST/resources/reports/UCR%20Summary%20of%20Reported%20Crimes%20in%20the%20Nation%202024.pdf.
8.	Kena and Thompson, Hate Crime Victimization, 2005-2019, p. 2.
9.	FBI, Criminal Justice Information Services Division, 2025.0 National Incident-Based Reporting System User Manual, June 23, 2025, pp. 103-105, https://le.fbi.gov/file-repository/nibrs-user-manual-2025-0-062625.pdf/view.
10.	Kena and Thompson, Hate Crime Victimization, 2005-2019, p. 5.
11.	FBI, 2025.0 National Incident-Based Reporting System User Manual, pp. 103-105.
12.	FBI, 2025.0 National Incident-Based Reporting System User Manual, p. 5.
13.	BJS, "National Crime Victimization Survey Data Dashboard," 2024, p. 2, https://ncvs.bjs.ojp.gov/methodology#vicdef.
14.	FBI, 2025.0 National Incident-Based Reporting System User Manual, pp. 7-10.
15.	Kena and Thompson, Hate Crime Victimization, 2005-2019, p. 6.
16.	FBI, Hate Crime Data Collection Guidelines and Training Manual, March 1, 2022, p. 3, https://le.fbi.gov/file-repository/hate-crime-data-collection-guidelines-and-training-manual.pdf/view.
17.	There was a period from about 2019 to 2021 when LEA participation in the UCR hate crime statistics program declined. This drop may have been partially due to the nationwide transition to the National Incident-Based Reporting System (NIBRS) happening at the time, which was associated with some disruptions in crime data reporting. For more information on NIBRS, see CRS Report R46668, The National Incident-Based Reporting System (NIBRS): Benefits and Issues, by Nathan Kemper.
18.	FBI, UCR Summary of Reported Crimes in the Nation, 2024, p. 18.
19.	Jacob Kaplan, Decoding FBI Crime Data (2025), chap. 9, https://ucrbook.com/hate_crimes.html.
20.	For an example of mandated statewide hate crime reporting standards, see Gurbir S. Grewal, Attorney General of New Jersey, "Attorney General's Bias Incident Investigation Standards," April 5, 2019, https://www.nj.gov/oag/dcj/agguide/Bias-Invest-Standards_040519.pdf.
21.	There could be many different reasons why there are so many zero report agencies despite the comparatively higher numbers of self-reported hate crime victimizations across the country each year. For more information about this topic, see CRS Report R46318, Federal Data on Hate Crimes in the United States, by Nathan James and Nathan Kemper.
22.	NCVS's methodology requires direct interviews with each person 12 or older in the household but makes exceptions in certain cases. Specifically, interview by proxy can be conducted when the individual is (1) a 12- or 13-year-old and a knowledgeable household member insists that he or she cannot be directly interviewed, (2) incapacitated, or (3) absent from the household during the entire field-interviewing period. For more information, see BJS, "Survey Methodology for Criminal Victimization in the United States," https://bjs.ojp.gov/sites/g/files/xyckuh236/files/media/document/ncvs_methodology.pdf.
23.	People living in group quarters such as dormitories and rooming houses are included in the sample.
24.	CRS used general variance function parameters and formulae provided by BJS to calculate standard errors for estimates. Estimates were flagged as unreliable if their relative standard errors (standard error/estimate*100) exceeded 50. For more information on the NCVS's generalized variance functions, see G. Lance Couzens et al., User's Guide to the National Crime Victimization Survey (NCVS) Generalized Variance Functions, BJS, https://bjs.ojp.gov/sites/g/files/xyckuh236/files/media/document/gvf_users_guide.pdf.
25.	Susannah N. Tapp and Emilie J. Coen, Criminal Victimization, 2024, BJS, September 2025, p. 4, https://bjs.ojp.gov/document/cv24.pdf.
26.	A single hate crime incident can have multiple offenses, victims, offenders, and bias motivations. For example, if a single incident was reported as having both an anti-Jewish and anti-lesbian motivation, that single incident would be counted under both the religion and sexual orientation categories. Conversely, a single hate crime incident can also have zero individual victims if it was, for example, destruction of property targeting a business, religious institution, or some other entity.
27.	Amy Farrell and Sarah Lockwood, "Addressing Hate Crime in the 21st Century: Trends, Threats, and Opportunities for Intervention," Annual Review of Criminology, vol. 6, no. 1 (2023), pp. 111-112.
28.	Michael Lieberman, "Hate Crimes, Explained," Southern Poverty Law Center, October 17, 2025, https://www.splcenter.org/resources/reports/hate-crimes-explained/.
29.	Crime rates were derived from FBI Summary Reporting System data, available at https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads.
30.	FBI, UCR Summary of Reported Crimes in the Nation, 2025, August 2026, p. 20, https://cde.ucr.cjis.gov/LATEST/resources/reports/UCR_Summary_of_Reported_Crimes_in_the_Nation_2025.pdf#page=20.
31.	There are two types of crimes in the NCVS: (1) person crimes (e.g., assault) and (2) household property crimes. Every year from 2015 to 2024, at least 86% of NCVS hate crimes were person crimes, so the analysis in this report focuses on them. However, bias-motivated household property crimes also generally increased from 2015 to 2024 (approximately 11.3 property crimes per 100,000 households in 2015 and 46.0 per 100,000 households in 2024).
32.	Victims could report more than one type of evidence for a single hate crime incident.
33.	Kena and Thompson, Hate Crime Victimization, 2005-2019, p. 4.
34.	Although not all states consider 18 the age of majority, for the purposes of this report a juvenile is any individual under age 18. The FBI began recording data on juvenile status of offenders and victims in 2013. See FBI, Hate Crime Statistics, 2024: Methodology, 2025, pp. 2-3, https://cde.ucr.cjis.gov/LATEST/webapp/#/pages/downloads.
35.	This section discusses NCVS data only, as UCR data do not provide information on victim sex.
36.	To explore these trends and others in the NCVS data, see the NCVS Dashboard (N-DASH) on BJS's website at https://ncvs.bjs.ojp.gov/multi-year-trends/characteristic.
37.	This section discusses NCVS data only, as UCR data do not provide information on victim race or ethnicity.
38.	Crimes against society include offenses such as prostitution or drug/narcotic offenses. For further information, see FBI, "NIBRS Offense Codes," 2011, pp. 1-4, https://ucr.fbi.gov/nibrs/2011/resources/nibrs-offense-codes.
39.	In this case, the category of assault offenses consists of intimidation, simple assault, and aggravated assault. For definitions of these different types of assault offenses, see FBI, "NIBRS Offense Definitions," 2019, p. 2, https://ucr.fbi.gov/nibrs/2018/resource-pages/nibrs_offense_definitions-2018.pdf#page=2.
40.	The NCVS data did not have an option for "Intimidation" for offense type.
41.	Simple assaults is defined by BJS as assaults without a weapon that resulted in no injury, minor injury, or an injury requiring less than two days of hospitalization. "Aggravated assaults" are assaults involving the use of a weapon or resulting in a serious injury to the victim. For more information, see BJS, "Terms and Definitions," https://ncvs.bjs.ojp.gov/terms#terms.
42.	There was a total of 46 different possible location types in the UCR data, most of which were relatively rare locations for hate crimes to occur, such as bars/nightclubs (1.7%), malls (0.4%), and amusement parks (0.1%), among others.
43.	These analyses were limited to single-offender incidents because, per the format of the current UCR data, only one offender's race can be recorded per incident, even when there are multiple offenders. This leaves ambiguity about the racial identity of all parties in the case of multi-offender incidents. Given that most hate crime incidents from 2015 to 2024 (60.0%) were recorded as having one offender, CRS decided to limit these particular offender-level analyses to avoid any ambiguous or potentially misleading results.
44.	FBI, Hate Crime Statistics, 2024: Methodology, p. 6.
45.	BJS, National Crime Victimization Survey: NCVS-2 Crime Incident Report, July 1, 2019, p. 14, https://bjs.ojp.gov/document/ncvs24_cir.pdf#page=14.
46.	Data are available at https://walker-data.com/tidycensus/.
47.	Other victim types (e.g., businesses, religious organizations) were excluded to make for a more analogous comparison to the ACS estimates.
48.	For this analysis, some race/ethnicity/ancestry bias motivations (e.g., anti-Arab) were excluded due to not being easily parsed out racial categories in the 2024 ACS. Furthermore, data were restricted to only single bias incidents.
49.	Pew Research Center, "2023-24 U.S. Religious Landscape Study Interactive Database," 2025, https://www.pewresearch.org/religious-landscape-study/about/.
50.	Other victim types (e.g., businesses, religious organizations) were excluded to make for a more analogous comparison to the ACS estimates.
51.	There was no way to infer specific bias motivations within the religion bias category in the NCVS data, as had been possible in the race/ethnicity/ancestry bias category. Thus, CRS used only the UCR victimization data for this analysis of religiously motivated crimes.
52.	Farrell and Lockwood, "Addressing Hate Crime in the 21st Century, p. 113.
53.	Mina Cikara et al., "Hate Crime Towards Minoritized Groups Increases as They Increase in Sized-Based Rank," Nature Human Behavior, vol. 6, no. 11 (2022), p. 1537.
54.	Joshua D. Freilich and Steven M. Chermak, Hate Crimes (Center for Problem-Oriented Policing, 2013), p. 13, https://popcenter.asu.edu/sites/g/files/litvpz3631/files/sites/default/files/problems/pdfs/hate_crimes.pdf.
55.	Heather Zaykowski et al., "Examining the Paradox of Crime Reporting: Are Disadvantaged Victims More Likely to Report to the Police?," Law and Society Review, vol. 53, no. 4 (2019), p. 1.
56.	FBI, Hate Crime Data Collection Guidelines and Training Manual, p. 6.
57.	FBI, UCR Summary of Reported Crimes in the Nation, 2024, pp. 23-27.
58.	Kateira Aryaeinejad and Thomas Leo Scherer, The Role of the Internet and Social Media on Radicalization: What Research Sponsored by the National Institute of Justice Tells Us, National Institute of Justice, April 2024, pp. 14-16, https://www.ojp.gov/pdffiles1/nij/305797.pdf#page=16.
59.	For more information on this program, see CRS In Focus IF10691, The Edward Byrne Memorial Justice Assistance Grant (JAG) Program, by Nathan James.
60.	OMB, "Revisions to OMB's Statistical Policy Directive No. 15: Standards for Maintaining, Collecting, and Presenting Federal Data on Race and Ethnicity," 89 Federal Register 22182-22196, March 29, 2024.
61.	Justia, "Hate Crime Laws: 50-State Survey," May 2024, https://www.justia.com/criminal/hate-crime-laws-50-state-survey/.
62.	See Federal Emergency Management Agency, "Nonprofit Security Grant Program," July 30, 2026, https://www.fema.gov/grants/preparedness/nonprofit-security.
63.	DOJ, "Matthew Shepard and James Byrd Jr. Hate Crimes Prevention Act," https://www.justice.gov/archives/hatecrimes/matthew-shepard-and-james-byrd-jr-hate-crimes-prevention-act.