Federal Data on Hate Crimes in the United States

Contents

• The Hate Crime Statistics Act
• Federal Hate Crime Data
• Hate Crime Statistics Program
• National Crime Victimization Survey
• Differences in the Two National Measures of Hate Crimes
• Are Hate Crimes Underreported to the FBI by Law Enforcement?
• Improving Hate Crime Data: Considerations for Policymakers

Summary

A relatively recent series of high-profile crimes where the offenders' actions appeared to be motivated by their bias or animosity towards a particular race, ethnicity, or religion might contribute to a perception that hate crimes are on the rise in the United States. These incidents might also generate interest among policymakers about how the federal government collects data on hate crimes committed in the United States.

The Federal Bureau of Investigation (FBI) started its Hate Crime Statistics program pursuant to the requirement in the Hate Crime Statistics Act (HSCA, P.L. 101-275) that the Department of Justice (DOJ) collect 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." In addition to the FBI's Hate Crime Statistics program, DOJ also collects data on hate crime victimizations through the Bureau of Justice Statistics' (BJS') National Crime Victimization Survey (NCVS). The NCVS measures self-reported criminal victimizations including those perceived by victims to be motivated by an offender's bias against them for belonging to or being associated with a group largely identified by the characteristics outlined in the HSCA.

Scholars, advocates, and members of the media have pointed out that there is a significant disparity between the number of hate crimes reported by the FBI each year and the number of hate crime victimizations reported by BJS. This has led some to criticize the hate crime data published by the FBI as an undercount of the number of hate crimes committed in the United States each year. However, this statistics gap can be partially explained by the different measures and methodologies utilized by the FBI and BJS to collect these data. For example, the FBI only reports on crimes that have been reported to the police, while BJS collects reports of criminal victimizations that may or may not meet the statutory definition of a hate crime and may or may not have been reported to the police. There are a number of reasons why some victims do not report their victimization to the police, including fear of reprisal, not wanting the offender to get in trouble, believing that police would not or could not do anything to help, and believing the crime to be a personal issue or too trivial to report.

There are also several reasons why a hate crime that was reported to the police might not be subsequently reported to the FBI for their Hate Crime Statistics program. Deciding whether a crime meets the statutory definition of a hate crime requires law enforcement agencies to investigate allegations of hate crime motivations before making a final determination. Reporting by law enforcement agencies to the FBI might be hampered by the fact that some law enforcement agencies do not have the training necessary to investigate potential bias-motivated offenses effectively. In addition, differing definitions between the FBI and state statutes as to what constitutes a hate crime generate confusion as to which standard should be used to determine whether a hate crime occurred and should be reported.

In 2021, the FBI plans to transition to the National Incident Based Reporting System (NIBRS) and will no longer accept non-NIBRS compliant data from law enforcement agencies. Policymakers might have an interest in how NIBRS differs from the FBI's current hate crime reporting program and whether full participation in NIBRS might improve the quality and completeness of federal hate crime data. However, like the FBI's current crime reporting program, participation in the NIBRS program is voluntary, and policymakers might consider steps Congress could take to promote wide-scale adoption of NIBRS.

---

The United States has experienced a series of high-profile violent crimes where the offenders' actions appeared to be motivated by their bias or animosity towards a particular race, ethnicity, or religion. For example, shootings at synagogues in Pittsburgh, PA, and Poway, CA; a driver speeding his car into protestors at a "Unite the Right" rally in Charlottesville, VA; a shooting at a Walmart in El Paso, TX, where the shooter allegedly said he was targeting "Mexicans" and espoused concerns about the "invasion" of the United States by immigrants; and reports of hate crimes against Asian-Americans during the Coronavirus pandemic contribute to a perception that hate crimes are on the rise in the United States. The salience of these events and how they are covered in the media might also contribute to the perception that there is a growing number of hate crimes (also known as bias crimes or bias-motivated offenses) being perpetrated in communities across the country.¹ Policymakers might turn to hate crime data collected by the Department of Justice (DOJ) to understand if there has actually been an increase in hate crimes in the United States and, if so, the nature of the increase. Policymakers might also utilize these same data to craft a policy response to hate crimes that is grounded in the data and conduct oversight of the federal government's efforts to combat these crimes.

This report begins with an overview of federal sources of data on hate crimes. This includes a brief overview of the Hate Crime Statistics Act (HCSA, P.L. 101-275), which requires DOJ to collect and report data on hate crimes, and the two systems DOJ employs to collect these data: the Federal Bureau of Investigation's (FBI's) Hate Crime Statistics Program and the Bureau of Justice Statistics' (BJS') National Crime Victimization Survey (NCVS). The report then discusses two salient issues regarding hate crime statistics: the large difference between the number of hate crimes reported by the FBI and the number of hate crime victimizations reported by BJS, and concerns about law enforcement agencies underreporting hate crimes to the FBI. The report concludes with a discussion of whether the wide-scale adoption of the FBI's National Incident Based Reporting System (NIBRS) might serve as a means of improving federal hate crime data.

The Hate Crime Statistics Act

The HCSA requires DOJ to collect 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." Congress required DOJ to collect these data because, at the time, few states collected data on hate crimes and there were no national data.² Policymakers believed that national data would reveal the scope of the problem and provide a basis for more effective law enforcement efforts to address hate crimes.³

Over the years since the HCSA was enacted, Congress has expanded the definition of what constitutes a hate crime for data collection purposes. The act initially required DOJ to collect data on hate crimes based on race, religion, sexual orientation, or ethnicity. In 2009, Congress amended the act to require DOJ to collect data on hate crimes based on the victims' gender or gender identity (P.L. 111-84) or disability (P.L. 103-322). P.L. 111-84 also required DOJ to collect and report data on hate crimes committed by and against juveniles.

The HCSA initially included a sunset provision that would have ended the requirement for DOJ to collect hate crime data after 1994. However, the Church Arson Prevention Act (P.L. 104-155) removed that provision.

Federal Hate Crime Data

To meet the requirements of the HCSA and subsequent amendments, DOJ collects and reports data on hate crimes that occur in the United States through two sources: the Hate Crime Statistics program and the NCVS.

Hate Crime Statistics Program

DOJ fulfills the HCSA's requirement by collecting supplemental data on hate crimes through the FBI's Uniform Crime Reporting (UCR) program. The Hate Crime Statistics Program collects data about hate crime offenders' bias motivations for the set of offenses already reported to the UCR program.⁴ Under the Hate Crime Statistics Program, the victim of a hate crime can be an individual, a business, an institution, or society as a whole.

Hate Crime Statistics Program data is collected and reported to the FBI by law enforcement agencies across the country. Agency participation in the Hate Crime Statistics Program, like the UCR program, is voluntary but most agencies participate. In 2018, more than 16,100 law enforcement agencies in all 50 states and the District of Columbia participated in the Hate Crime Statistics Program.⁵ The agencies that participated represented jurisdictions that include nearly 307 million people.⁶ For a point of comparison, in 2008 there were a reported 17,985 state and local law enforcement agencies that employed at least one full-time officer or the equivalent in part-time officers.⁷

The FBI requires law enforcement agencies to use a two-step process for investigating hate crimes before reporting them to the Hate Crime Statistics Program.⁸ In the first step, the law enforcement officer that initially responds to a potential hate crime incident is responsible for determining whether there is any indication that the offense was motivated by bias against an individual's perceived membership in one of the groups specified in the HCSA. If there is an indication of a bias motivation, the incident is designated as a suspected bias-motivated crime and forwarded to an investigator. In the second step, the investigator is responsible for reviewing the facts of the incident and making the final determination as to whether the crime meets the HCSA definition of a hate crime. According to the FBI, an agency should only report an incident as a hate crime when a law enforcement investigation reveals sufficient evidence to lead a reasonable and prudent person to conclude that the offender's actions were motivated, in whole or in part, by his or her bias.⁹

Law enforcement agencies can submit data on single and multiple bias incidents. Single bias incidents are those in which one or more of the offenses committed during an incident are motivated by the same bias. Multiple bias incidents are those in which one or more of the offenses committed during an incident are motivated by two or more biases.

Annual hate crime data published by the FBI differs from traditional UCR crime data published by the FBI in an important way. For most crimes, the FBI estimates full-year crime data for law enforcement agencies that submit less than 12 months of data to the UCR. In contrast, hate crime data published by the FBI only includes offenses reported by the police; no estimation for missing data is done by the FBI for the Hate Crime Statistics Program.

National Crime Victimization Survey

BJS has collected data on hate crime victimizations through the NCVS since 2003.¹⁰ The NCVS data is collected through annual interviews with residents of a nationally representative sample of households.¹¹ All people age 12 or older in the sampled households are interviewed. The NCVS collects self-reported data on non-fatal personal crime victimizations (sexual assault, robbery, aggravated and simple assaults, and personal larceny) and property crime victimizations (burglary, motor vehicle theft, and other thefts) regardless of whether the crimes were reported to the police.

The NCVS uses the same HCSA definition of a hate crime as the FBI. The NCVS collects data on crimes that victims perceive to be motivated by an offender's bias against them based on their race, gender and gender identity, religion, disability, sexual orientation, or ethnicity. Hate crime victimizations are counts of "a single victim or household that experienced a criminal incident believed by the victim to be motivated by hate."¹² In the NCVS data, hate crime victimizations for personal crimes are counts of individual victims, while hate crime victimizations for property crimes are counts of victimized households.

In order for a victimization to be classified as a hate crime in the NCVS, the victim has to report one of three types of evidence of the offender's bias: (1) the offender used hate language, (2) the offender left hate signs or symbols at the scene, or (3) police investigators confirmed that a hate crime occurred.¹³

Table 1 compares the methodologies of the UCR Hate Crime Statistics Program and the NCVS.

Differences in the Two National Measures of Hate Crimes

A perennial issue that can cause confusion for those unfamiliar with the FBI's and BJS's data collection goals and methodologies is the difference between the number of hate crime incidents reported by the FBI and the number of hate crime victimizations reported by BJS. For example, for 2018 (the most recent data available) the FBI reported that there were approximately 7,100 hate crime incidents that involved approximately 8,800 victims.¹⁴ In comparison, BJS reported that there were an estimated 198,000 hate crime victimizations in 2017.¹⁵

What might explain the difference in the two national measures of hate crimes? The answer lies partially in the fact that the data reported by the FBI and BJS reflect different goals for collecting data on hate crimes. The FBI data only reflect hate crime incidents that are reported to law enforcement, and where law enforcement concludes that a hate crime has occurred and reports it to the FBI's Hate Crime Statistics Program. In contrast, the goal of the NCVS hate crime data collection effort is to estimate the total number of hate crime victimizations that occur each year, including victimizations that are not reported to law enforcement agencies (i.e., a portion of the dark figure¹⁶ of crime). Because the NCVS collects data on reported and unreported hate crime victimizations, its totals will always be larger than the FBI's hate crime data.

Another explanation for the difference between the two measures are the different standards needed to be met to be counted as a hate crime in the FBI's Hate Crime Statistics Program and the NCVS. For a hate crime to be counted by the FBI, law enforcement must have sufficient evidence that would lead a reasonable and prudent person to conclude that the offender's actions were motivated, in whole or in part, by his or her bias. In contrast, under the NCVS, an incident is counted as a hate crime if the victim believes that the offense was based on their race, ethnicity, religion, disability, sexual orientation, gender, or gender identity, and the offender used hate language, hate symbols, or a law enforcement investigation concluded that a hate crime had occurred. An independent investigation of the perceived bias is not necessary in every case for the NCVS interviewers to include the offense as a hate crime.

The goals and methodologies described above help explain why the NCVS estimates of hate crime victimizations are higher than the number of hate crime incidents reported by the FBI. At the same time, the FBI's Hate Crime Statistics Program collects data on a larger number of victim types and crimes that may be motivated by the offender's bias than the NCVS. For example, the FBI collects data on bias motivated homicides and vandalisms, which are not be captured by the NCVS. Law enforcement agencies can also report data on hate crimes against individuals, businesses, religious institutions, other institutions, and society as a whole to the FBI, whereas the NCVS only collects data on hate crimes against individuals (i.e., personal crimes) and households (i.e., property crimes).

Are Hate Crimes Underreported to the FBI by Law Enforcement?

A common criticism of the FBI's hate crime data is that a large proportion of participating law enforcement agencies report zero hate crimes in a given year (zero-reporting agencies), leading some advocacy groups to accuse the zero-reporting agencies of underreporting hate crimes.¹⁷ The evidence presented to support these accusations are discrepancies between hate crime figures reported by the FBI and the self-reported hate crime figures tabulated by community organizations serving the communities that are often the targets of hate crime (e.g., organizations serving the LGBTQ, Jewish, Muslim, or Arab communities).

Research suggests that some law enforcement agencies have underreported the number of hate crime incidents to the FBI. In one study, researchers reviewed a sample of assault incident reports from seven local law enforcement agencies across the country that were not classified as hate crimes to see if there was any indication that the offenses had a bias motivation.¹⁸ Incidents where there was a clear indication that bias was a predominant motivating factor in the assault were coded as bias-motivated, and other incidents were coded as ambiguous if there was an indication of bias but also evidence of some other identifiable triggering event or alternative motivation. The study found that for some of the incidents, there was evidence that they were motivated by the alleged perpetrator's bias, but that these misclassification errors were relatively infrequent and varied by law enforcement agency. The estimated proportion of misclassified cases for each agency ranged from zero to 8% of all assault incidents when both bias-motivated and ambiguous incidents were considered and from zero to 3% when only bias-motivated cases were considered. While the proportion of misclassified assault cases for any individual agency is relatively low, if the percentage of misclassified cases reported in this study was generalizable to the universe of all assaults, it would account for thousands of hate crimes that were not reported to the Hate Crime Statistics Program.

Another study of the accuracy of hate crime reporting utilized incident-based crime data (see discussion of expanding the National Incident Based Reporting System, below) from four local law enforcement agencies to evaluate whether hate crimes were being misclassified.¹⁹ This study looked at all criminal incidents, not just assaults, reported to the four agencies in 2008 and examined not only whether hate crimes were misclassified as non-bias-motivated offenses, but also whether non-bias-motivated offenses were wrongly classified as hate crimes and how these errors compared to misclassification errors for other non-hate crimes. This study found that undercounting of hate crimes was the most common misclassification error in the records they examined. The researchers noted that "extending error rates to the population suggest that the estimated number of bias crimes that go unaccounted is noticeable."²⁰

Even though the research described above did not focus on local law enforcement agencies who reported zero hate crimes, it is these agencies in particular that critics argue are likely to have underreported hate crimes. As shown in Figure 1, the vast majority of agencies that participate in the Hate Crime Statistics Program are zero-reporting agencies, leading critics to assume that hate crimes are significantly underreported to the FBI.

In order for a law enforcement agency to be considered a "participant," it has to submit data on the number of hate crimes for at least part of the year or a letter signed by the police chief certifying that no hate crimes occurred that year in its jurisdiction.²¹ From 1996 to 2017, at least 80% of Hate Crime Statistics Program participating law enforcement agencies in any given year reported zero hate crimes. The proportion of participating law enforcement agencies that were zero-reporting agencies generally increased from 2001 to 2014. There was a slight decrease in this proportion after 2014, but in 2018 nearly 9 out of 10 participating law enforcement agencies reported zero hate crimes.

Figure 1. Number of Law Enforcement Agencies Participating in the FBI's Hate Crime Statistics Program, 1996-2018

Source: U.S. Department of Justice, Federal Bureau of Investigation, Hate Crime Statistics for 1996-2018.

Aside from misclassification errors, there are several reasons that might explain why a law enforcement agency does not report any hate crimes in a given year. The first, and most straightforward, reason is because no hate crimes occurred. Given that law enforcement agency jurisdictions include communities with as little as a few hundred residents, it is not implausible that some residents, especially those that live in very small and homogeneous communities, did not experience any hate crimes. Second, in order for a law enforcement agency to report a hate crime to the FBI, it must be reported to the police. Data from the NCVS indicates that on average, half of hate crime victimizations were not reported to the police from 2013 to 2017.²² Hate crime victims might choose not to report the incident to the police for a variety of reasons, including

• fear of retaliation,
• embarrassment that they were victimized,
• a belief that the crime was not motivated by the perpetrator's bias,
• lack of familiarity with a state's hate crime laws,
• distrust of law enforcement,
• a belief that law enforcement will not investigate the case,
• fear of being exposed as a member of the LGBTQ community, or
• fear of being re-traumatized by the criminal justice system.²³

Even when a hate crime is reported to state and local law enforcement, an investigation must be conducted into the perceived bias to determine if the offense was bias-motivated before reporting it to the FBI as a hate crime. This step can be challenging for law enforcement agencies, especially small agencies with relatively few resources. When there is evidence that a hate crime might have occurred, law enforcement agencies have to complete additional investigative steps to determine whether an offense meets the statutory definition of a hate crime, and in some cases law enforcement officers might not be trained sufficiently on recognizing biases in crimes to conduct such investigations.²⁴ Few states provide mandatory training for law enforcement officers on investigating, identifying, and reporting hate crimes, and in the states that do, there is little oversight to confirm that law enforcement officers are receiving the training and applying it correctly.²⁵

Ambiguity in the circumstances surrounding hate crimes can also lead to an undercounting. Under the Hate Crime Statistics Program, law enforcement agencies report the number of hate crimes that were "motivated in whole or in part by bias." Law enforcement officers might have difficulty applying this standard in cases where a bias motivation might not be obvious, especially when considering hate crimes that were motivated "in part" by an offender's bias.²⁶ While a cross burning on the front yard of a black family's home is an unambiguous hate crime, in other cases the motivation of the alleged perpetrators might not be so clear. These ambiguous hate crimes can be classified into two categories: response/retaliation events and target-selection events.²⁷

• Response/retaliation events are those where the offense was first triggered by something other than bias, but at some point bias exacerbates the incident into a hate crime.²⁸ For example, a white motorist and a black motorist get into a dispute because their cars were involved in an accident. However, after a few minutes, the white motorist assaults the black motorist while yelling racial slurs. In this case, the incident was not initiated because of the white motorist's bias against the black motorist, but the white motorist's bias eventually resulted in him assaulting the black motorist.
• Target-selection events are those where a target of a crime is selected because of the offender's bias against members of the group, but the offender's bias in not obvious.²⁹ For example, someone might rob men leaving bars that are known to be frequented by same sex couples because the offender believes they will be less likely to report the offense because they might not want to be identified as being a member of the LGBTQ community.

In addition to issues related to law enforcement officer training on identifying hate crimes for submission to the FBI, differences in how a hate crime is defined under state law and under the HCSA can create its own ambiguities. For example, gender identity is a protected class under the HCSA, but it might not be a recognized bias motivation under a state's laws. As such, if a law enforcement officer is more familiar with the state's hate crime definition, he or she might not identify an offense based on gender-bias as a potential hate crime. As one group of researchers noted:

Even when potential bias crimes are reported to a participating agency, the agency must then recognize any indications of bias, determine whether the incident is bias motivated, document the motivation, and submit the incident to UCR. Empirical evidence suggests that the processing of bias-crime reporting across participating law enforcement agencies is variable and subject to much error and interpretation by local departments.³⁰

Improving Hate Crime Data: Considerations for Policymakers

Congress passed the HCSA with the intent of collecting national data on bias-motivated offenses that could be used to inform federal hate crime policy. While DOJ has taken steps to collect these data, the hate crime data reported by the FBI is incomplete and the NCVS self-reported hate crime victimization data likely includes incidents that would not meet the legal standard needed to be charged as hate crime. Hate crime data "missing" from the FBI's Hate Crime Statistics program results from a series of complications associated with collecting these data (e.g., victims might not report the offense to the police, law enforcement agencies might fail to correctly identify potential hate crimes, or law enforcement agencies might not routinely and systematically report hate crime data to the FBI). Policymakers may have an interest in what steps Congress could take to help improve the quality of the FBI's hate crime data. One option on the horizon might be the wide-scale adoption of the National Incident Based Reporting System (NIBRS).

The FBI is in the process of phasing out the UCR summary reporting system and having all law enforcement agencies submit data through NIBRS. The FBI reports that it will begin collecting only NIBRS-compliant data from law enforcement agencies starting on January 1, 2021.³¹ To support state and local law enforcement agencies' transitions to NIBRS, state and local governments that are not certified as NIBRS compliant have been required since FY2018 to use 3% of their award under the Edward Byrne Memorial Justice Assistance Grant (JAG) program to achieve compliance.³²

Compared to the UCR summary reporting system, NIBRS collects more data on a wider variety of offenses.³³ NIBRS asks participating law enforcement agencies to collect and report incident-level data on offenders, victims, the relationship between victims and offenders, and the circumstances surrounding the incident for 52 different offenses. In comparison, the current summary reporting system is largely a tabulation of the number of eight Part I offenses reported to the police.³⁴

As a part of NIBRS, reporting agencies can identify whether an offense was motivated by an offender's bias against the victim for each reported offense. Under the Hate Crime Statistics Program, law enforcement agencies that are not currently submitting NIBRS-compliant data submit a supplemental summary report to the FBI when there is evidence that one or more crimes in their jurisdiction involved a bias motivation. It has been argued that hate crime reporting will increase as more agencies adopt NIBRS because reporting the presence or absence of bias motivations is built into NIBRS.³⁵ In addition to making it easier for law enforcement agencies to report hate crimes to the FBI, NIBRS provides data on a wider variety of offenses, including those that were motivated by offenders' bias against their victims, and data on the context of hate crimes (e.g., locations where hate crimes occur, the relationship between alleged perpetrators and victims of hate crimes, whether alleged offenders are residents of the community where they committed their offenses, the weapons used in the offenses (if any), and the types and seriousness of injuries sustained by hate crime victims).³⁶

While the FBI might stop accepting crime data from non-NIBRS compliant law enforcement agencies next year, participation in the program is still voluntary. If a law enforcement agency does not believe it is worth the time and effort to adopt NIBRS and the state does not mandate that it participates in the program, there is no federal mandate or incentive for the agency to participate. Therefore, policymakers might have an interest in what steps Congress could take to promote wide-scale adoption of the program. Congress could consider placing a condition on a program such as JAG that would require law enforcement agencies to submit NIBRS data to the FBI or face a penalty under the program. However, the JAG program already provides a financial incentive to participate fully in the FBI's crime reporting program. Half of a state's allocation is based on its proportion of the average number of violent crimes reported in the United States over the past three years, and allocations for local governments are based on their proportion of the average number of violent crimes reported in the state over the past three years.³⁷ The Bureau of Justice Assistance reports that NIBRS data will be used to calculate JAG awards once NIBRS replaces the summary reporting system.³⁸ In addition, in order for local governments to be eligible for a direct award under the program, they have to have submitted violent crime data for 3 of the past 10 years.³⁹ Yet, even with these incentives some law enforcement agencies in the United States do not participate in the UCR because compiling the data can be difficult and time consuming, and many small agencies might not have the resources needed to fully comply with the FBI's data collection and submission requirements. Thus, Congress could also consider authorizing a new grant program that would provide funding to state and local governments to cover expenses related to transitioning to NIBRS, such as purchasing new software and computers, or training officers on how to use NIBRS.

While NIBRS might provide some administrative efficiency with regard to reporting hate crimes, it does not address some of the other issues law enforcement agencies currently have with reporting hate crimes through the UCR program. Implementing NIBRS does not address hate crime victims being reluctant to report an offense to the police, the need for training for law enforcement officers on how to identify potential hate crimes, or the need to improve law enforcement agencies processes for investigating potential hate crimes, nor will it resolve differences between the HCSA and state hate crime definitions.

Author Contact Information

Footnotes

1.	For example, research suggests that increased media coverage of school shootings after a high-profile incident can lead people to believe that school shootings occur with more frequency than they actually do. See Glenn W. Muschert, "Research in School Shootings," Sociology Compass, vol. 1, no. 1 (2007), pp. 60-80; and H. Jaymi Elsass, Jaclyn Schidkraut, and Mark C. Stafford, "Studying School Shootings: Challenges and Considerations for Research," American Journal of Criminal Justice, vol. 14 (2016), pp. 444-464.
2.	U.S. Congress, House Committee on the Judiciary, Hate Crime Statistics Act, report to accompany H.R. 1048, 101st Cong., 1st sess., June 23, 1989, H.Rept. 101-109, p. 2; U.S. Congress, Senate Committee on the Judiciary, Hate Crime Statistics Act, report to accompany S. 419, 101st Cong., 1st sess., May 1, 1989, S.Rept. 101-21, p. 3.
3.	Ibid.
4.	These offenses include crimes against persons (homicide, rape, robbery, aggravated and simple assaults, intimidation, human trafficking, and involuntary servitude), crimes against property (robbery, burglary, larceny-theft, motor vehicle theft, arson, and destruction/damage/vandalism), and crimes against society (drugs or narcotics offenses, gambling offenses, prostitution offenses, and weapons law violations).
5.	U.S. Department of Justice, Federal Bureau of Investigation, Hate Crime Statistics, 2018 (hereinafter, "Hate Crime Statistics, 2018").
6.	Ibid.
7.	The number of state and local law enforcement agencies in 2008 is the most recent figure published by BJS. Another census of state and local law enforcement agencies was conducted by BJS in 2014, but figures from that census are not yet available. Brian A. Reaves, Census of State and Local Law Enforcement Agencies, U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, NCJ 233982, Washington, DC, July 2011, p. 2.
8.	U.S. Department of Justice, Federal Bureau of Investigation, Hate Crimes Data Collection Guidelines and Training Manual, version 2.0, February 27, 2015, pp. 2-3.
9.	Hate Crime Statistics, 2018, Methodology.
10.	Madeline Masucci and Lynn Langton, Hate Crime Victimization, 2004–2015, U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, NCJ 250653, Washington, DC, June 2017, p. 1 (hereinafter "BJS, Hate Crime Victimizations").
11.	Ibid., p. 9.
12.	Ibid., p. 2.
13.	Ibid., p. 10.
14.	Hate Crime Statistics, 2018, Table 1.
15.	Barbara Oudekerk, "Hate Crime Statistics," briefing prepared for the Virginia Advisory Committee, U.S. Commission on Civil Rights, U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Statistics, March 29, 2019 (hereinafter, "Oudekerk, 'Hate Crime Statistics'").
16.	The dark figure of crime refers to crimes that are not reported to the police or are undetected.
17.	See, for example, Muslim Advocates, "Muslim Advocates Responds to Rise in Annual Hate Crimes Data, Despite Rampant Underreporting," press release, November 14, 2018, https://muslimadvocates.org/2018/11/muslim-advocates-responds-to-rise-in-annual-hate-crimes-data-despite-rampant-underreporting/; Human Right Campaign, "New Statistics Show Alarming Increase in Number of Reported Hate Crimes," November 13, 2018, https://www.hrc.org/blog/new-fbi-statistics-show-alarming-increase-in-number-of-reported-hate-crimes; and Arjun Singh Sethi, "The FBI Recorded a Surge of Hate Crimes Last Year. But it Undercounted—By a Lot," Washington Post, November 14, 2018.
18.	Jack McDevitt, James Cronin, and Jennifer Balboni et al., Bridging the Information Disconnect in National Reporting of Bias Crime, Research in Brief, The Center for Criminal Justice Policy Research, Northeastern University, Boston, MA, p. 5.
19.	James J. Nolan, Stephen M. Haas, and Erica Turley et al., "Assessing the 'Statistical Accuracy' of the National Incident-Based Reporting System Hate Crime Data," American Behavioral Scientist, vol. 59, no. 12 (2015), pp. 1562-1587 (hereinafter, "Nolan, et al., 'Assessing the "Statistical Accuracy" of the National Incident-Based Reporting System Hate Crime Data'").
20.	Ibid., p. 1582.
21.	Stephen M. Haas, James J. Jordan, and Erica Turley et al., Assessing the Validity of Hate Crime Reporting: An Analysis of NIBRS Data, State of West Virginia, Department of Military Affairs and Public Safety, Division of Justice and Community Services, Office of Research and Strategic Planning, Charleston, WV, July 2011, p. 4.
22.	Oudekerk, "Hate Crime Statistics."
23.	Harbani Ahuja, "The Vicious Cycle of Hate: Systemic Flaws in Hate Crime Documentation in the United States and the Impact on Minority Communities," Cardozo Law Review, vol. 37, no. 5 (June 2016), pp. 1882-1883 (hereinafter, "Ahuja, 'The Vicious Cycle of Hate'").
24.	William B. Rubenstein, "The Real Story of U.S. Hate Crime Statistics: An Empirical Analysis," Tulane Law Review, vol. 78, no. 4 (March 2004), p. 1220 (hereinafter, "Rubenstein, 'The Real Story of U.S. Hate Crime Statistics'").
25.	Ahuja, "The Vicious Cycle of Hate," p. 1892.
26.	Nolan, et al., "Assessing the 'Statistical Accuracy' of the National Incident-Based Reporting System Hate Crime Data," p. 1564.
27.	Ibid.
28.	Ibid., pp. 1564-1565.
29.	Ibid., p. 1565.
30.	Shea W. Cronin, Jack McDevitt, and Amy Farrell et al., "Bias-Crime Reporting: Organizational Responses to Ambiguity, Uncertainty, and Infrequency in Eight Police Departments," American Behavioral Scientist, vol. 51, no. 2 (October 2007), p. 216.
31.	U.S. Department of Justice, Federal Bureau of Investigation, National Incident Based Reporting System (NIBRS), https://www.fbi.gov/services/cjis/ucr/nibrs.
32.	U.S. Department of Justice, Office of Justice Programs, Bureau of Justice Assistance, Edward Byrne Memorial Justice Assistance Grant (JAG) Program Frequently Asked Questions (FAQs), updated March 2019, p. 4 (hereinafter, "JAG FAQs").
33.	For more information, see U.S. Department of Justice, Federal Bureau of Investigation, SRS to NIBRS: The Path to Better UCR Data, March 28, 2017, https://www.fbi.gov/services/cjis/cjis-link/srs-to-nibrs-the-path-to-better-ucr-data.
34.	The eight Part I offenses are homicide, rape, robbery, aggravated assault, burglary, larceny-theft, motor vehicle theft, and arson.
35.	James J. Nolan III, Yoshio Akiyama, and Samuel Berhanu, "The Hate Crime Statistics Act of 1990: Developing a Method for Measuring the Occurrence of Hate Violence," American Behavioral Scientist, vol. 46, no. 1 (September 2002), p. 146.
36.	Ibid., p. 147.
37.	For more information on how allocations are calculated under the JAG program see CRS In Focus IF10691, The Edward Byrne Memorial Justice Assistance Grant (JAG) Program.
38.	JAG FAQs, p. 4.
39.	34 U.S.C. §10156(e)(3).