In 2023, 8,842 murderers in the United States were white, while 6,405 were Black. A further 461 murderers were of another race, including American Indian or Alaska Native, Asian, and Native Hawaiian or Other Pacific Islander. However, not all law enforcement agencies submitted homicide data to the FBI in 2023, meaning there may be more murder offenders of each race than depicted. While the majority of circumstances behind murders in the U.S. are unknown, narcotics, robberies, and gang killings are most commonly identified.

In 1980, the National Institute of Justice awarded a grant to the Cornell University College of Human Ecology for the establishment of the Center for the Study of Race, Crime, and Social Policy in Oakland, California. This center mounted a long-term research project that sought to explain the wide variation in crime statistics by race and ethnicity. Using information from eight ethnic communities in Oakland, California, representing working- and middle-class Black, White, Chinese, and Hispanic groups, as well as additional data from Oakland's justice systems and local organizations, the center conducted empirical research to describe the criminalization process and to explore the relationship between race and crime. The differences in observed patterns and levels of crime were analyzed in terms of: (1) the abilities of local ethnic communities to contribute to, resist, neutralize, or otherwise affect the criminalization of its members, (2) the impacts of criminal justice policies on ethnic communities and their members, and (3) the cumulative impacts of criminal justice agency decisions on the processing of individuals in the system. Administrative records data were gathered from two sources, the Alameda County Criminal Oriented Records Production System (CORPUS) (Part 1) and the Oakland District Attorney Legal Information System (DALITE) (Part 2). In addition to collecting administrative data, the researchers also surveyed residents (Part 3), police officers (Part 4), and public defenders and district attorneys (Part 5). The eight study areas included a middle- and low-income pair of census tracts for each of the four racial/ethnic groups: white, Black, Hispanic, and Asian. Part 1, Criminal Oriented Records Production System (CORPUS) Data, contains information on offenders' most serious felony and misdemeanor arrests, dispositions, offense codes, bail arrangements, fines, jail terms, and pleas for both current and prior arrests in Alameda County. Demographic variables include age, sex, race, and marital status. Variables in Part 2, District Attorney Legal Information System (DALITE) Data, include current and prior charges, days from offense to charge, disposition, and arrest, plea agreement conditions, final results from both municipal court and superior court, sentence outcomes, date and outcome of arraignment, disposition, and sentence, number and type of enhancements, numbers of convictions, mistrials, acquittals, insanity pleas, and dismissals, and factors that determined the prison term. For Part 3, Oakland Community Crime Survey Data, researchers interviewed 1,930 Oakland residents from eight communities. Information was gathered from community residents on the quality of schools, shopping, and transportation in their neighborhoods, the neighborhood's racial composition, neighborhood problems, such as noise, abandoned buildings, and drugs, level of crime in the neighborhood, chances of being victimized, how respondents would describe certain types of criminals in terms of age, race, education, and work history, community involvement, crime prevention measures, the performance of the police, judges, and attorneys, victimization experiences, and fear of certain types of crimes. Demographic variables include age, sex, race, and family status. For Part 4, Oakland Police Department Survey Data, Oakland County police officers were asked about why they joined the police force, how they perceived their role, aspects of a good and a bad police officer, why they believed crime was down, and how they would describe certain beats in terms of drug availability, crime rates, socioeconomic status, number of juveniles, potential for violence, residential versus commercial, and degree of danger. Officers were also asked about problems particular neighborhoods were experiencing, strategies for reducing crime, difficulties in doing police work well, and work conditions. Demographic variables include age, sex, race, marital status, level of education, and years on the force. In Part 5, Public Defender/District Attorney Survey Data, public defenders and district attorneys were queried regarding which offenses were increasing most rapidly in Oakland, and they were asked to rank certain offenses in terms of seriousness. Respondents were also asked about the public's influence on criminal justice agencies and on the performance of certain criminal justice agencies. Respondents were presented with a list of crimes and asked how typical these offenses were and what factors influenced their decisions about such cases (e.g., intent, motive, evidence, behavior, prior history, injury or loss, substance abuse, emotional trauma). Other variables measured how often and under what circumstances the public defender and client and the public defender and the district attorney agreed on the case, defendant characteristics in terms of who should not be put on the stand, the effects of Proposition 8, public defender and district attorney plea guidelines, attorney discretion, and advantageous and disadvantageous characteristics of a defendant. Demographic variables include age, sex, race, marital status, religion, years of experience, and area of responsibility.

In 2023, the FBI reported that there were 9,284 Black murder victims in the United States and 7,289 white murder victims. In comparison, there were 554 murder victims of unknown race and 586 victims of another race. Victims of inequality? In recent years, the role of racial inequality in violent crimes such as robberies, assaults, and homicides has gained public attention. In particular, the issue of police brutality has led to increasing attention following the murder of George Floyd, an African American who was killed by a Minneapolis police officer. Studies show that the rate of fatal police shootings for Black Americans was more than double the rate reported of other races. Crime reporting National crime data in the United States is based off the Federal Bureau of Investigation’s new crime reporting system, which requires law enforcement agencies to self-report their data in detail. Due to the recent implementation of this system, less crime data has been reported, with some states such as Delaware and Pennsylvania declining to report any data to the FBI at all in the last few years, suggesting that the Bureau's data may not fully reflect accurate information on crime in the United States.

Between 1982 and September 2024, 82 out of the 151 mass shootings in the United States were carried out by White shooters. By comparison, the perpetrator was African American in 26 mass shootings, and Latino in 12. When calculated as percentages, this amounts to 54 percent, 17 percent, and eight percent respectively. Race of mass shooters reflects the U.S. population Broadly speaking, the racial distribution of mass shootings mirrors the racial distribution of the U.S. population as a whole. While a superficial comparison of the statistics seems to suggest African American shooters are over-represented and Latino shooters underrepresented, the fact that the shooter’s race is unclear in around nine percent of cases, along with the different time frames over which these statistics are calculated, means no such conclusions should be drawn. Conversely, looking at the mass shootings in the United States by gender clearly demonstrates that the majority of mass shootings are carried out by men. Mass shootings and mental health With no clear patterns between the socio-economic or cultural background of mass shooters, increasing attention has been placed on mental health. Analysis of the factors Americans considered to be to blame for mass shootings showed 80 percent of people felt the inability of the mental health system to recognize those who pose a danger to others was a significant factor. This concern is not without merit – in over half of the mass shootings since 1982, the shooter showed prior signs of mental health issues, suggesting improved mental health services may help deal with this horrific problem. Mass shootings and guns In the wake of multiple mass shootings, critics have sought to look beyond the issues of shooter identification and their influences by focusing on their access to guns. The majority of mass shootings in the U.S. involve firearms which were obtained legally, reflecting the easy ability of Americans to purchase and carry deadly weapons in public. Gun control takes on a particular significance when the uniquely American phenomenon of school shootings is considered. The annual number of incidents involving firearms at K-12 schools in the U.S. was over 100 in each year since 2018. Conversely, similar incidents in other developed countries exceptionally rare, with only five school shootings in G7 countries other than the U.S. between 2009 and 2018.

Black people were over twice as likely to be arrested as white people – there were 20.4 arrests for every 1,000 black people, and 9.4 for every 1,000 white people.

Incident-based crime statistics (actual incidents, rate per 100,000 population, percentage change in rate, unfounded incidents, percent unfounded, total cleared, cleared by charge, cleared otherwise, persons charged, adults charged, youth charged / not charged), by detailed violations (violent, property, traffic, drugs, other Federal Statutes), police services in Ontario, 1998 to 2024.

Incident-based crime statistics (actual incidents, rate per 100,000 population, percentage change in rate, unfounded incidents, percent unfounded, total cleared, cleared by charge, cleared otherwise, persons charged, adults charged, youth charged / not charged), by detailed violations (violent, property, traffic, drugs, other Federal Statutes), police services in Alberta, 1998 to 2024.

In 2023, the violent crime rate in the United States was 363.8 cases per 100,000 of the population. Even though the violent crime rate has been decreasing since 1990, the United States tops the ranking of countries with the most prisoners. In addition, due to the FBI's transition to a new crime reporting system in which law enforcement agencies voluntarily submit crime reports, data may not accurately reflect the total number of crimes committed in recent years. Reported violent crime rate in the United States The United States Federal Bureau of Investigation tracks the rate of reported violent crimes per 100,000 U.S. inhabitants. In the timeline above, rates are shown starting in 1990. The rate of reported violent crime has fallen since a high of 758.20 reported crimes in 1991 to a low of 363.6 reported violent crimes in 2014. In 2023, there were around 1.22 million violent crimes reported to the FBI in the United States. This number can be compared to the total number of property crimes, roughly 6.41 million that year. Of violent crimes in 2023, aggravated assaults were the most common offenses in the United States, while homicide offenses were the least common. Law enforcement officers and crime clearance Though the violent crime rate was down in 2013, the number of law enforcement officers also fell. Between 2005 and 2009, the number of law enforcement officers in the United States rose from around 673,100 to 708,800. However, since 2009, the number of officers fell to a low of 626,900 officers in 2013. The number of law enforcement officers has since grown, reaching 720,652 in 2023. In 2023, the crime clearance rate in the U.S. was highest for murder and non-negligent manslaughter charges, with around 57.8 percent of murders being solved by investigators and a suspect being charged with the crime. Additionally, roughly 46.1 percent of aggravated assaults were cleared in that year. A statistics report on violent crime in the U.S. can be found here.

The purpose of the study was to assess the impact of Latino ethnicity on pretrial release decisions in large urban counties. The study examined two questions: Are Latino defendants less likely to receive pretrial releases than non-Latino defendants? Are Latino defendants in counties where the Latino population is rapidly increasing less likely to receive pretrial releases than Latino defendants in counties where the Latino population is not rapidly increasing? The study utilized the State Court Processing Statistics (SCPS) Database (see STATE COURT PROCESSING STATISTICS, 1990-2004: FELONY DEFENDANTS IN LARGE URBAN COUNTIES [ICPSR 2038]). The SCPS collects data on felony cases filed in state courts in 40 of the nation's 75 largest counties over selected sample dates in the month of May of every even numbered year, and tracks a representative sample of felony case defendants from arrest through sentencing. Data in the collection include 118,556 cases. Researchers supplemented the SCPS with county-level information from several sources: Federal Bureau of Investigation Uniform Crime Reporting Program county-level data series of index crimes reported to the police for the years 1988-2004 (see UNIFORM CRIME REPORTS: COUNTY-LEVEL DETAILED ARREST AND OFFENSE DATA, 1998 [ICPSR 9335], UNIFORM CRIME REPORTING PROGRAM DATA [UNITED STATES]: COUNTY-LEVEL DETAILED ARREST AND OFFENSE DATA, 1990 [ICPSR 9785], 1992 [ICPSR 6316], 1994 [ICPSR 6669], 1996 [ICPSR 2389], 1998 [ICPSR 2910], 2000 [ICPRS 3451], 2002 [ICPSR 4009], and 2004 [ICPSR 4466]). Bureau of Justice Statistics Annual Survey of Jails, Jurisdiction-Level data series for the years 1988-2004 (see ANNUAL SURVEY OF JAILS: JURISDICTION-LEVEL DATA, 1990 [ICPSR 9569], 1992 [ICPSR 6395], 1994 [ICPSR 6538], 1996 [ICPSR 6856], 1998 [ICPSR 2682], 2000 [ICPSR 3882], 2002 [ICPSR 4428], and 2004 [ICPSR 20200]). Bureau of Justice Statistics National Prosecutors Survey/Census data series 1990-2005 (see NATIONAL PROSECUTORS SURVEY, 1990 [ICPSR 9579], 1992 [ICPSR 6273], 1994 [ICPSR 6785], 1996 [ICPSR 2433], 2001 census [ICPSR 3418], and 2005 [ICPSR 4600]). United States Census Bureau State and County Quickfacts. National Center for State Courts, State Court Organization reports, 1993 (see NCJ 148346), 1998 (see NCJ 178932), and 2004 (see NCJ 212351). Bureau of Justice Statistics Felony Defendants in Large Urban Counties reports, 1992 (see NCJ 148826), 1994 (see NCJ 164616), 1996 (see NCJ 176981), 1998 (see NJC 187232), 2000 (see NCJ 202021), and 2002 (see NJC 210818). The data include defendant level variables such as most serious current offense charge, number of charges, prior felony convictions, prior misdemeanor convictions, prior incarcerations, criminal justice status at arrest, prior failure to appear, age, gender, ethnicity, and race. County level variables include region, crime rate, two year change in crime rate, caseload rate, jail capacity, two year change in jail capacity, judicial selection by election or appointment, prosecutor screens cases, and annual expenditure on prosecutor's office. Racial threat stimuli variables include natural log of the percentage of the county population that is Latino, natural log of the percentage of the county population that is African American, change in the percentage of the county population that is Latino over the last six years and change in the percentage of the county population that is African American over the last six years. Cross-level interaction variables include percentage minority (Latino/African American) population zero percent to 15 percent, percentage minority (Latino/African American) population 16 percent to 30 percent, and percentage minority (Latino/African American) population 31 percent or higher.

Incident-based crime statistics (actual incidents, rate per 100,000 population, percentage change in rate, unfounded incidents, percent unfounded, total cleared, cleared by charge, cleared otherwise, persons charged, adults charged, youth charged / not charged), by detailed violations (violent, property, traffic, drugs, other Federal Statutes), police services in Manitoba, 1998 to 2024.

Incident-based crime statistics (actual incidents, rate per 100,000 population, percentage change in rate, unfounded incidents, percent unfounded, total cleared, cleared by charge, cleared otherwise, persons charged, adults charged, youth charged / not charged), by detailed violations (violent, property, traffic, drugs, other Federal Statutes), police services in British Columbia, 1998 to 2024.

This study examined the ways in which the model of the Kings County Felony Domestic Violence Court (FDVC) changed the way cases were processed and adjudicated, the impact of this approach on outcomes, and its effects on recidivism. In order to evaluate the implementation and effectiveness of the FDVC, the researchers selected three samples of cases for collection of detailed data and comparisons on case characteristics, processing, and outcomes. First, felony domestic violence cases indicted from 1995 to early 1996 before the FDVC was established, and adjudicated by various parts of the state Supreme Court were studied. These pre-FDVC cases provided a comparison group for assessing differences associated with the FDVC model. Very few of these cases had felony protection order violations as the sole or top indictment charge, since they predated the implementation of the expanded criminal contempt law that went into effect in September 1996. Second, a sample of cases adjudicated by FDVC in its early period (the first half of 1997, after the model was fully implemented) and similar in indictment charges to the pre-FDVC cases was selected. These were cases that had indictment charges other than, or in addition to, felony criminal contempt charges for protection order violations. In other words, these were cases that would have been indicted and adjudicated in the state Supreme Court even without application of the September 1996 law. Third, because the September 1996 law felonizing many protection order violations (under criminal contempt statutes) broadened the types of cases handled by the Supreme Court, compared with those handled in the Supreme Court prior to this law, an additional sample of cases adjudicated by the FDVC (beginning in the first half of 1997) was selected. This was a small sample in which felony protection order violations were the only indicted felony charges. These cases would not have been indicted on felonies during the pre-FDVC period, and so would have remained in the criminal courts as misdemeanors. The inclusion of this sample allowed the researchers to assess how the protection order violation cases were different from the general population of FDVC cases, and how they might be handled differently by the Court and partner agencies. These cases were designated "CC-only" because their only felony indictment was for criminal contempt, the law under which felony protection order violations were charged. Variables in Part 1, Recidivism Data, contain information on number of appearance warrants issued, days incarcerated for predisposition, number of appearances for predisposition and post-disposition, bail conditions (i.e., batterer treatment or drug treatment), top charge at arrest, indictment, and disposition, indications of defendant's substance abuse of alcohol, marijuana, or other drugs, and psychological problems, types of disposition and probation, months of incarceration, sentence conditions, history of abuse by defendant against the victim, length of abuse in months, history of physical assault and sexual abuse, past weapon use, and medical attention needed for past domestic violence. Additional variables focus on whether an order of protection was issued before the current incident, whether the defendant was arrested for past domestic violence with this victim, total number of known victims, weapon used during incident, injury during the incident, medical attention sought, number of final orders of protection, whether the defendant was jailed throughout the pending case, number of releases during the case, number of reincarcerations after release, whether the victim lived with the defendant, whether the victim lived with children in common with the defendant, relationship between the victim and the defendant, number of months the victim had known the defendant, number of children in common with the defendant, whether the victim attempted to drop charges, whether the victim testified at trial, whether a victim advocate was assigned, total violations during pending case, predisposition violations, and number of probation violations. Demographic variables in Part 1 include defendant and victims' gender, race, victim age at defendant's arrest, defendant's income, employment status, and education. Variables in Part 2, Top Charge Data, relating to the defendant include number and types of prior arrests and convictions, top charge at arrest, severity of top charge at arrest, top charge at grand jury indictment, severity of top charge indictment, disposition details, Uniform Crime Reporting (UCR) arrest indicators, child victim conviction indicator, drug conviction indicator, weapon conviction indicator, types of probation, sentence, disposition, and offenses. Demographic variables in Part 2 include sex and race of the defendant.

By 1996 it became apparent that the Los Angeles county jails faced a serious overcrowding problem. Two possible solutions to the problem were to build more jail capacity or to divert a greater number of incoming inmates to community-based, intermediate sanctions. The research team for this study was asked to review a 1996 profile of inmates in the Los Angeles jail system and to determine how many of them might have been good candidates for intermediate sanctions such as electronic monitoring, work release, house arrest, and intensive supervision. The researchers selected a sample of 1,000 pre-adjudicated (or unconvicted) inmates from the total census of inmates in jail custody on January 15, 1996, to study in more detail. Of the 1,000 offenders, the researchers were able to obtain jail and recidivism data for two years for 931 inmates. For each of these offenders, information on their prior criminal history, current offense, and subsequent recidivism behavior was obtained from official records maintained by several county agencies, including pretrial services, sheriff's department, probation, and courts. Demographic variables include date of birth, race, and gender. Prior criminal history variables for each prior adult arrest include type of filing charge, case disposition, type of sentence and sentence length imposed, and total number of prior juvenile petitions sustained. Current offense variables include arrest date, crime type for current arrest, crime charge, type and date of final case disposition, and sentence type and length, if convicted. Strike information collected includes number of strikes and the offense that qualified as a strike. Jail custody variables include the jail entry and exit data for the current offense and the reason for release, if released. Lastly, two-year follow-up variables include the date, type, and disposition of each subsequent arrest between January 15, 1996, and January 15, 1998.

Incident-based crime statistics (actual incidents, rate per 100,000 population, percentage change in rate, unfounded incidents, percent unfounded, total cleared, cleared by charge, cleared otherwise, persons charged, adults charged, youth charged / not charged), by detailed violations (violent, property, traffic, drugs, other Federal Statutes), police services in Saskatchewan, 1998 to 2024.

Effective October 1, 1994, the state of North Carolina implemented a new structured sentencing law that applied to all felony and misdemeanor crimes (except for driving while impaired) committed on or after October 1, 1994. Under the new structured sentencing law parole was eliminated, and a sentencing commission developed recommended ranges of punishment for offense and offender categories, set priorities for the use of correctional resources, and developed a model to estimate correctional populations. This study sought to investigate sentencing reforms by looking at the effects of structured sentencing on multiple aspects of the adjudication process in North Carolina. A further objective was to determine whether there were differences in the commission of institutional infractions between inmates sentenced to North Carolina prisons under the pre-structured versus structured sentencing laws. Researchers hoped that the results of this study may help North Carolina and jurisdictions around the country (1) anticipate the likely effects of structured sentencing laws, (2) design new laws that might better achieve the jurisdictions' goals, and (3) improve the potential of sentencing legislation in order to enhance public safety in an effective and equitable way. Administrative records data were collected from two sources. First, in order to examine the effects of structured sentencing on the adjudication process in North Carolina, criminal case data were obtained from the North Carolina Administrative Office of the Courts (Parts 1 and 2). The pre-structured sentencing and structured sentencing samples were selected at the case level, and each record in Parts 1 and 2 represents a charged offense processed in either the North Carolina Superior or District Court. Second, inmate records data were collected from administrative records provided by the North Carolina Department of Correction (Part 3). These data were used to compare the involvement in infractions of inmates sentenced under both pre-structured and structured sentencing. The data for Part 3 focused on inmates entering the prison system between June 1, 1995, and January 31, 1998. Variables for Parts 1 and 2 include type of charge, charged offense date, method of disposition (e.g., dismissal, withdrawal, jury trial), defendant's plea, verdict for the offense, and whether the offense was processed through the North Carolina Superior or District Court. Structured sentencing offense class and modified Uniform Crime Reporting code for both charged and convicted offenses are presented for Parts 1 and 2. There are also county, prosecutorial district, and defendant episode identifiers in both parts. Variables related to defendant episodes include types of offenses within episode, total number of charges and convictions, whether all charges were dismissed, whether any felony charge resulted in a jury trial, and the adjudication time for all charges. Demographic variables for Parts 1 and 2 include the defendant's age, race, and gender. Part 3 variables include the date of prison admission, sentence type, number of prior incarcerations, number of years served during prior incarcerations, maximum sentence length for current incarceration, jail credit in years, count of all infractions during current and prior incarcerations, reason for incarceration, infraction rate, the risk for alcohol and drug dependency based on alcohol and chemical dependency screening scores, and the number of assault, drug/alcohol, profanity/disobedience, work absence, and money/property infractions during an inmate's current incarceration. Demographic variables for Part 3 include race, gender, and age at the time of each inmate's prison admission.

California reported the largest number of homicides to the FBI in 2023, at 1,929 for the year. Texas recorded the second-highest number of murders, with 1,845 for the year. Homicide victim demographics There were a total of 19,252 reported homicide cases in the U.S. in 2023. When looking at murder victims by gender and ethnicity, the vast majority were male, while just over half of the victims were Black or African American. In addition, homicide victims in the United States were found most likely to be between the ages of 20 and 34 years old, with the majority of victims aged between 17 to 54 years old. Are murders up? In short, no – since the 1990s the number of murders in the U.S. has decreased significantly. In 1990, the murder rate per 100,000 people stood at 9.4, and stood at 5.7 in 2023. It should be noted though that the number of homicides increased slightly from 2014 to 2017, although figures declined again in 2018 and 2019, before ticking up once more in 2020 and 2021. Despite this decline, when viewed in international comparison, the U.S. murder rate is still notably high. For example, the Canadian homicide rate stood at 1.94 in 2023, while the homicide rate in England and Wales was even lower.

The number of reported cases of rape in Denmark increased significantly over the past years. While *** rapes were reported in 2013, the number peaked in 2021, with over ***** cases. The number of rape charges also increased during this period. In 2022, ***** persons were charged for rape, meaning that a person was charged in ** percent of the rape cases that year. The increasing number of reported cases of rape may not necessarily be down to an increase in the actual number, but is probably also related to an increased awareness around the topic.

The crime rate in the United Kingdom was highest in England and Wales in 2024/25, at **** crimes per 1,000 people, compared with Scotland, which had **** crimes per 1,000 population, and Northern Ireland, at **** crimes per 1,000 people in 2023/24. During this time period, the crime rate of England and Wales has usually been the highest in the UK, while Scotland's crime rate has declined the most, falling from **** crimes per 1,000 people in 2002/03, to just **** by 2021/22. Overall crime on the rise In 2024/25, there were approximately **** million crimes recorded by the police in England and Wales. Although this was a slight decline on the previous two years, it was still far higher than during the mid-2010s. While crime declined quite significantly between 2002/03 and 2013/14, this trend went into reverse in subsequent years. While there are no easy explanations for the recent uptick in crime, it is possible that reduced government spending on the police service was at least partly to blame. In 2009/10, for example, government spending on the police stood at around **** billion pounds, with this cut to *****billion in 2013/14. One of the most visible consequences of these cuts was a sharp reduction in the number of police officers in the UK. Between 2010 and 2017, the number of officers fell by 20,000, although the number of officers returned to pre-austerity levels by the 2020s. A creaking justice system During the period of austerity, the Ministry of Justice as a whole saw its budget sharply decline, from *** billion pounds in 2009/10 to just **** billion by 2015/16. Although there has been a reversal of the cuts to budgets and personnel in the justice system, the COVID-19 pandemic hit the depleted service hard in 2020. A backlog of cases grew rapidly, putting a strain on the ability of the justice system to process cases quickly. In 2022, for example, it took on average *** days for a crown court case to go from offence to conclusion, compared with *** days in 2018. There is also the issue of overcrowding in prisons, with the number of prisoners in England and Wales dangerously close to operational capacity in recent years.

In 2022, the incarceration rate of African Americans in local jails in the United States was *** incarcerations per 100,000 of the population -- the highest rate of any race or ethnicity. The second-highest incarceration rate was among American Indians/Alaska Natives, at *** incarcerations per 100,000 of the population.