Weapon Use And Violent Crime
Weapon Use and Violent Crime
National Crime Victimization Survey, 1993-2001
By Craig Perkins
For nonfatal violent crimes, offenders were more likely to
have a firearm than a knife or club. From 1993 to 2001 the
rate of firearm violence fell 63%
- Approximately half of all robberies, about a quarter of
all assaults, and roughly a twelfth of all rapes/sexual
assaults involved an armed assailant. About 90% of
homicide victims were killed with a weapon.
- Firearm violence rates for blacks age 12 or older (8.4
per 1,000 blacks) were 40% higher than rates for Hispanics
200% higher than rates for whites (2.8 per
- Blacks were about 9 times more likely than whites to be
murdered with a firearm.
- On average black victims of firearm violence were 3
years younger than white victims -- 29 versus 32.
- From 1993 through 2001 blacks accounted for 46% of
homicide victims and 54% of victims of firearm homicide but
12% of the U.S. population.
- The likelihood of an injury was the same for victims
facing armed and unarmed offenders (26%); serious injury
was more likely from armed offenders (7% versus 2%).
- From 1993 through 2001 the number of murders declined
36% while the number of murders by firearms dropped 41%.
- From 1994 through 1999, the years for which data are
available, about 7 in 10 murders at school involved some
type of firearm, and approximately 1 in 2 murders at
school involved a handgun.
Estimates from the National Crime Victimization Survey
(NCVS)indicate that between 1993 and 2001 approximately
26% of the average annual 8.9 million violent victimizations
were committed by offenders armed with a weapon. About 10%,
or 846,950 victimizations each year, involved a firearm.
From 1993 through 2001 violent crime declined 54%; weapon
violence went down 59%; and firearm violence, 63%.
Males, blacks and Hispanics, the young, and those with the
lowest annual household income were more vulnerable to
weapon violence in general and firearm violence in
particular than their respective counterparts.
For the 9-year period beginning with 1993, 23% of white
victims of violence and 36% of black victims were victims
of violence involving an offender armed with a weapon.
About 7% of white victims and 17% of black victims were
involved in incidents in which an offender was armed with
Forty-five percent of all violence with a weapon involved
victims between ages 25 and 49, and 38% involved victims
between ages 15 and 24.
Blacks were about 9 times more likely than whites to be
victims of gun-related homicides (25 per 100,000 blacks
age 12 or older versus 3 per 100,000 whites.)
While victimizations involving knives comprised 6% of all
violent crimes resulting in an injury, these victimizations
accounted for about 24% of all serious injuries experienced
by crime victims.
The most common locales for armed violence and gun violence
were the streets: those away from the victim' home (30% of
violence with a weapon and 35% of gun violence) and those at
or near the victim's home (27% of armed violence and 25% of
Most violence involving a weapon and most firearm violence
occurred while the victims were engaged in leisure activities
away from home (27% and 27%)and commuting to work (23% and 25%,
Weapon use varied by type of crime. Offenders had weapons
in about half of robberies, a fourth of assaults, and a
twelfth of rapes/sexual assaults.
Weapons and violent crime
On average each year between 1993 and 2001, approximately
26% (about 2.3 million) of the estimated 8.9 million violent
crimes in the United States were committed by offenders
armed with guns, knives, or objects used as weapons.
Firearm violence accounted for 10% of all violent crimes;
about 6% were committed with a knife or other sharp object
such as scissors, ice pick, or broken bottle; 4% with blunt
objects such as a brick, bat, or bottle; and 5% were
committed with unspecified/ "other" objects used as weapons.
The National Crime Victimization Survey (NCVS)
The NCVS is the Nation's primary source of information on
criminal victimization. Data are continuously obtained
from a nationally representative sample of approximately
43,000 households comprising nearly 80,000 persons age 12
or older. Household members are asked about the frequency,
characteristics, and consequences of victimization.
The survey enables the Bureau of Justice Statistics (BJS)
to estimate the rate of victimization for rape, attempted
rape, sexual assault, robbery, assault, theft, household
burglary, and motor vehicle theft. The rates describe the
vulnerability to crime by the population as a whole as well
as by segments of the population such as women, the elderly,
members of racial and ethnic groups, and city dwellers.
For the most current estimates of criminal victimization in
the United States, see Criminal Victimization 2001: Changes
2000-2001 with Trends 1993-2001 <www.ojp.usdoj.gov/bjs>.
Definitions of weapons
Firearms include handguns (pistols, revolvers, derringers)
and shotguns, rifles, and other firearms (excluding BB and
pellet guns and air rifles).
Sharp objects include knives and other sharp edged and/or
pointed objects (scissors, ice picks, and axes).
Blunt objects include rocks, clubs, blackjacks, bats, and
Other weapons include ropes, chains, poison, martial arts
weapons, BB guns, and objects that could not be classified.
Between 1993 and 2001 victims were confronted by offenders
armed with guns in about 27% of robberies, 8% of assaults,
and 3% of all rapes/sexual assaults.
On average, each year U.S. residents were victims of crimes
committed with firearms at a rate of 4 crimes per 1,000
persons age 12 or older. Of the average 847,000 violent
victimizations committed with firearms, about 7 out of 8
were committed with handguns.
Knives and sharp objects
Annually during the 9-year period, about 570,000 violent
victimizations were committed with a knife or other sharp
object, accounting for 6% of all violent crimes. Thirteen
percent of robberies, 6% of assaults, and 3% of rapes
were committed with a knife or other sharp object.
From 1993 through 2001 crimes involving knives or sharp
objects were committed at an average annual rate of 3 per
1,000 persons age 12 or older. In 85% of these
victimizations, about 480,000 each year, the weapon was a
knife. In the remainder, about 88,000 victimizations per
year, the weapon was another type of sharp object.
Armed with blunt objects such as bats, sticks, rocks,
clubs, or blackjacks, offenders committed approximately
356,000 violent crimes each year from 1993 through 2001.
Crimes by offenders armed with blunt objects were committed
at an average annual rate of 2 per 1,000 persons age 12 or
Between 1993 and 2001 about 5% of all violent crimes were
committed with weapons other than guns, knives, or blunt
objects. Such weapons include ropes, chains, poison, martial
arts weapons, BB guns (not considered to be firearms by the
NCVS), and objects that could not be classified.
Weapon use and crime outcome
Weapon use varied by crime. Robberies, followed by all
assaults, were more likely to involve an armed assailant
while rape/sexual assault was the least likely.
Armed robberies were more likely to be completed, resulting
in loss of property, than unarmed robberies. Higher
completion percentages occurred for robberies committed
with firearms than for robberies with knives and other
sharp objects or blunt objects/other weapons. Robberies
committed with knives and unarmed robberies were completed
at similar percentages.
All assaults in this report represent simple and aggravated
assault examined together.
Simple assault is an attack without a weapon resulting in
either no injury or minor injury.
Aggravated assault is an attack or attempted attack with a
weapon, regardless of whether an injury occurred and attack
without a weapon when serious injury results.
A discussion of minor and serious injury appears on
For the 9-year period beginning in 1993, the percentage of
rapes/sexual assaults that was completed did not vary
significantly depending on the offenders' possession of a
weapon. About 71% of rapes/sexual assaults involving no
weapon were completed; of such assaults with a weapon, 67%
The outcome for incidents of assault is measured by whether
a victim sustained an injury as a result of the crime.
Overall, about 1 in 4 assault victims were injured during
the incident. Victims of firearm violence were less likely
than other victims to be injured. About 1 in 3 assault
victims were injured when the offender possessed a blunt
object or some unspecified type of weapon. About 1 in 4
victims were injured when the assailant had a knife.
Males, blacks, Hispanics, and those between ages 15 and 24
were more vulnerable than their respective counterparts
to violent crime committed by armed assailants.
Between 1993 and 2001, approximately 1 in 3 male victims
of violent crime faced an armed offender. About 1 in 5
female victims of violent crime faced an armed assailant.
Males were twice as likely as females to be confronted by
an armed offender (14 versus 7 per 1,000 persons respectively).
This pattern was generally consistent across
weapon types; for each type of weapon, the victimization
rate for males was about twice that for females.
Offender use of firearms
Of incidents involving offenders with firearms,
* were shot (3%)
* were shot at but not hit (8%)
* were struck with a firearm (4%)
* were threatened with a firearm (72%)
* did not describe offender's use of firearms
Race and ethnicity
For each type of weapon, victimization rates for whites were
lower than those for blacks or Hispanics.***Footnote 1: In
this report race and ethnicity are analyzed together. White,
black, Asian/Pacific Islander, and American Indian refer to
non-Hispanic persons.*** Blacks were victimized by offenders
armed with guns at higher rates than Hispanics but at similar
rates as American Indians. Blacks had similar victimization
rates as Hispanics for crimes committed with knives or blunt
The rate of firearm violence for blacks was more than twice
that for whites (8 versus 3 per 1,000). The rate for Hispanics
(6 per 1,000)was about twice that for whites.
No significant differences separated the rates at which whites
and blacks were victimized by unarmed offenders.
The rate of armed violence for American Indians (25 per
1,000 American Indians) was--
- 43% higher than the rate for blacks (18 per 1,000 blacks)
- 78% higher than the rate for Hispanics (14 per 1,000
- 184% higher than the rate for whites (9 per 1,000).
Age, weapons, and violence
For overall violence, persons age 12-14, 15-17, and 18-20
were victimized at similar rates, higher than those for
persons age 21 or older. Vulnerability to victimization
by an armed offender similarly varied by the age of
victim. Younger persons, particularly those age 18-20, had
higher rates of victimization by armed offenders.
The rate of firearm violence was also highest for persons
age 18-20. Their rate (12 per 1,000 persons) was about 40%
higher than the rate for persons ages 15 to 17 and 21 to
Except for victims age 12-14, for whom firearm violence
constituted about 3% of all violent crime, firearm violence
accounted for between 9% and 13% of all violent crime for
each age group.
Similarly, crimes committed with knives/sharp objects
accounted for 6% to 8%, and crimes with other weapons,
9% to 11% of all violent crime for each age group
Forty-five percent of all armed violence involved victims
between 25 and 49, and 38% was against victims age 15- 24.
Race, ethnicity, and age
When race and ethnicity are added to the considerations of
age, some differences in victimization rates between age
categories emerge for crimes committed with any weapon
or with a firearm. Both for crimes committed with any
weapon and for firearm violence specifically, differences
between black and Hispanic victimization rates were
greatest within the age category 18-20.
Whites and blacks, age 18-20, were more likely than whites
and blacks of other ages to have been victims of weapon
violence in general and firearm violence in particular.
Hispanics of ages 15-17 and 18-20were more vulnerable
than other Hispanics to violence involving a weapon and
violence involving a firearm.
The rates of violent victimization, violent victimization
involving a weapon, and violent victimization involving a
firearm for persons age 18-20 were approximately 20 times
those of persons age 65 or older.
For blacks, whites, and Hispanics, victims of violent crime
were, on average, younger than the general population. For
the general population as well as for the victim population,
the mean age of whites was greater than the mean age of
Overall among the victims, blacks were older than Hispanics,
the youngest racial or ethnic group considered. This pattern
of relative ages was true for victims of violence involving
Annual household income
Persons with annual household incomes of less than $7,500
experienced both armed violence and firearm violence at
about 3 times the rates of persons with annual household
incomes of $50,000 or more (23.1 versus 7.3 armed
victimizations per 1,000 persons, and 8.4 versus 2.4
firearm victimizations, respectively).
Annual household income and race/ethnicity
Blacks at every income level were more vulnerable than
whites to be victims of firearm violence. Whites and
blacks with household incomes below $7,500 were more
vulnerable to armed violence than their counterparts
with higher incomes.
For violence by an offender with a weapon and for violence
by an offender armed with a gun, blacks with household
incomes of less than $50,000 were victimized at rates
higher than those of Hispanics or whites with similar
Weapons and injuries
For 1993-2001 about a quarter of violent crimes overall
resulted in an injury to the victim. Crimes committed
with weapons and crimes committed without weapons were
about equally likely to result in victim injury (26%).
Crimes committed with weapons, however, were about 3.5
times as likely to result in serious injury as crimes
committed by unarmed offenders (7% versus 2%,
Of all violence with a weapon, the crimes committed with
blunt objects/other weapons were the most often associated
with victim injury (36%). Twenty-eight percent of the
crimes with knives/sharp objects and 15% of crimes with
firearms involved injury.
Offenders armed with knives accounted for 6% of all violence
but 24% of all serious injuries having inflicted serious
injury on about 1 in 8 of their victims. About 1 in 15 victims
of offenders using a blunt object/other weapon and 1 in 22
victims of offenders with a firearm sustained serious injury.
Because completed rape is considered an injury, victims of
rape/sexual assault were more likely than robbery or assault
victims to be injured, regardless of offender weapon use.
For all weapon types, robbery victims were more likely than
assault victims to sustain injury.
Definition of types of injuries
Serious injuries include gunshot or knife wounds, broken
bones, loss of teeth, internal injuries, loss of
consciousness, and undetermined injuries requiring 2 or
more days of hospitalization.
Minor injuries include bruises, black eyes, cuts, scratches,
swelling, chipped teeth, and undetermined injuries requiring
less than 2 days of hospitalization.
Rape is sexual intercourse forced on the victim through
physical or psychological coercion. Forced sexual
intercourse means vaginal, anal, or oral penetration by
the offender(s), including penetration by a foreign
object. Victims can be male or female, and the rape can be
heterosexual or homosexual. NCVS defines completed rape as
a serious injury.
Rape without additional injuries, for the purposes of this
report, were categorized as an injury but not as a serious
or minor injury. Cases in which the victim suffered
additional injuries were grouped according to the severity
of those additional injuries.
Robbery and injuries
About half of victims of robbery by offenders armed with
blunt objects/ other weapons sustained an injury during
About a third of victims of robbery by unarmed offenders
(36%) and offenders armed with knives or sharp objects
(31%)sustained injury during the victimization.
Offenders armed with any weapon other than a firearm
inflicted a serious injury during about 1 in 7 robberies
that they committed.
Victims of robbery by offenders armed with blunt objects/
other weapons were more likely than victims of robbery by
offenders armed with a firearm to be attacked without a
Victims were injured in a third of all assaults by offenders
armed with blunt objects/other weapons. Less than a third of
assaults by offenders armed with guns or knives resulted
in injury (13% and 26%, respectively). Victims of
offenders armed with a knife or sharp object were the most
likely to sustain a serious injury; 12% of such offenses
resulted in serious injury.
About half of all victims of rape/sexual assault committed
by unarmed offenders were injured, compared to threequarters
of victims of such crimes by armed offenders.
***Footnote 2: The small number of sample cases of rape/
sexual assault, the least frequent nonlethal violent crime,
prevents further examination of injuries by type of
Injuries sustained include completed rape. (See definitions
of type of injuries on page 6.)
Characteristics of the incident
Time of incident
Violent crimes at night were more likely than crimes
occurring during the day to involve a weapon (30% versus
21%, respectively ) or a firearm (12% versus 6%,
respectively). Three of every five crimes committed by
an offender with a firearm occurred at night.
Activity and location
Crime by armed offenders was most likely to occur while the
victim was engaged in leisure activity away from home (27%)
or traveling to or from work or school (23%). Nearly 21% of
victims of armed violence were involved in some activity at
home at the time of the incident.
The most common location for crimes by armed offenders was
on the street away from the victim's home (30%). About a
quarter of all violence by armed offenders occurred at or
near the victim's home.
About 6% of armed violence, and 2% of firearm violence
occurred at a school or on school grounds.
Crimes committed by intimates were less likely than crimes
committed by strangers to involve a weapon. The offender
was armed in a third of all violence by a stranger and in
a sixth of all violence committed by an intimate.
Victims of crimes by strangers were also more likely than
victims of crimes by intimates to be confronted by an
offender with a firearm (14% versus 5%, respectively).
Violent victimization rates declined from 1993 to 2001.
Rates for crimes committed with firearms reflected a
larger decrease than did the rates for overall violence
and armed violence in general. Between 1993 and 2001
overall violence decreased 54%, armed violence fell 59%,
and firearm violence declined 63%.
The rates of firearm violence for blacks and Hispanics
fell relatively more than the rate for whites, 1993-2001.
In 1993 blacks and Hispanics were victims of firearm
violence at a rate of 13 firearm crimes per 1,000 persons,
about 3 times the rate for whites. By 2001 the rate for
blacks had fallen to about 4 per 1,000, roughly 2.5 times
that for whites. The rate for Hispanics in 2001 was less
than 3 victims of firearm violence per 1,000 Hispanics
in the population.
From 1993 to 2001, rates of violence involving firearms
declined among all age groups. The decline was greatest
among the youngest victims: by 2001, persons age 12-14
had experienced a 97% decrease in the rate of firearm
violence, and those age 15-17, a 77% decrease.
Homicide and weapons
Between 1993 and 2001 local law enforcement agencies
reported 160,396 murders and nonnegligent manslaughters
of persons age 12 or older to the FBI's Supplementary
Homicide Reports. (The homicides that occurred
in the events of September 11, 2001, were not included.)
There was an annual average of 17,822 murders of persons
age 12 or older. A weapon was used in 91% of these crimes.
In 4% of the homicides, the offender used a means such
as strangling, punching, and kicking. Information about
the weapon used was unavailable in 5% of all homicides.
Seventy percent of homicide victims were killed with a
firearm. Handguns were used in 56% of all homicides.
Male homicide victims were more likely than female victims
to have been killed by a firearm. Eight in ten
male homicide victims were killed with a firearm,
compared to 6 in 10 female victims.
Knives or other sharp objects were the second most frequently
used weapon in homicides of both males and females (13% and 22%,
respectively). Males were 5 times more likely than females
to be a victim of a homicide committed with a firearm (10
versus 2 per 100,000 persons, respectively), and twice as
likely to be murdered with a knife or other sharp object
(2 versus 1 per 100,000, respectively).
From 1993 to 2001, blacks were 12% of the U.S. population
age 12 or older but 46% of all homicide victims and 54% of
all victims of firearm homicide. Among homicide victims,
blacks were more likely than whites to have been killed
with a firearm. About 8 in 10 black homicide victims and
7 in 10 white homicide victims died from gunshot injuries.
Blacks were about 7 times more likely than whites to be a
homicide victim (30 versus 4 per 100,000 persons age 12 or
older respectively), and approximately 9 times more likely
to be a victim of a homicide committed with a firearm (25
versus 3 per 100,000 persons age 12 or older, respectively).
Persons age 18 to 24 were victims of homicide overall as
well as firearm homicides at the highest rates. Those age
18 to 24 were 3.5 times as likely as persons age 12
to 17 and about 3 times as likely as those age 25 to 64
to be killed with a firearm.
There appears to be a relationship between the age of the
victim and the type of homicide weapon. Victims of firearm
murders were, on average, 6 years younger than victims of
homicides committed with knives/other sharp objects (31
versus 37 years old, respectively), and 12 years younger
than victims of homicides committed with blunt objects
Murders of children under age 12
Weapons in general and firearms specifically were less
commonly used against murder victims under age 12 than
against those age 12 or older.
From 1993 through 2001, about 2 in 5 murders of children
under age 12 involved a weapon, and about 1 in 6 involved
a firearm. Among victims age 12 or older, 91% of the
murders were committed with a weapon, and 70% were
committed with a firearm.
In 1993 firearms accounted for 19% of murders of persons
under age 12. This peak of 195 incidents preceded a decline;
in 2001, 15% of all such murders (127 of the 824 total)
involved a firearm.
Race and age
On average, black murder victims were 5 years younger
than white victims (age 31 versus age 36, respectively).
The same age difference existed for firearm murders (age
28 versus age 33, respectively). Black and white victims
of murders committed with knives or blunt objects were,
on average, closer in age.
Trends in homicides, 1993-2001
The number of homicides declined 36% between 1993 and
2001. Firearmhomicides decreased 41% during the period.
In 1993, 72% of homicides of persons age 12 or older
were committed with firearms. In 2001, 66% were
committed with firearms.
Weapons, juveniles, and school violence
Annually on average, 1993 to 2001, about 703,800 violent
crimes against persons age 12-17 occurred at school or
on school property. About 8% of these crimes were
committed with a weapon; about 1% with a firearm.
About a third of armed assaults in schools resulted in
injury to the victim.
Between 1994 and 1999, the most recent year of available
data,172 homicides of both students and nonstudents took
place at school or on school property in the United
Of these, 69% were committed with a firearm. Eighteen
percent were committed with a knife or other sharp object.
Of the firearm homicides at school, three-quarters were
committed with handguns.
Between 1993 and 2001, about 61% of all victims of violent
crime reported taking a self-defensive measure during the
Most used nonaggressive means, such as trying to escape,
getting help, or attempting to scare off or warn the
offender. About 13% of victims of violent crime tried to
attack or threaten the offender. About 2% of victims of
violent crime used a weapon to defend themselves; half of
these, about 1% of violent crime victims, brandished a
Except for homicide data obtained from the FBI's Uniform
Crime Reporting Program, this report presents data from
the BJS National Crime Victimization Survey (NCVS).
Between 1993 and 2001 the Census Bureau interviewed
approximately 651,750 individuals age 12 or older in
336,295 households. For the NCVS data presented, response
rates varied between 93% and 96% of eligible households,
and between 89% and 92% of eligible individuals.
Violent acts covered in this report include murder, rape,
sexual assault, robbery, and assault (aggravated and
simple) against persons age 12 or older. Overall violent
crime is a combination of each type of crime.
Hypothesis testing and the NCVS
Standard error computations
Comparisons of estimates discussed in this report were
tested to determine if the differences were statistically
significant. Differences described as higher, lower, or
different passed a hypothesis test at the .05-level of
statistical significance (95%-level of confidence). That
is, the tested difference was greater than about twice
the standard error of that difference. Comparisons of
estimates statistically significant at the 0.10 level
(90%-confidence level), have differences described as
somewhat, marginal, or slight.
Caution is required when making comparisons of estimates
not explicitly discussed in this report. What may appear
to be a large difference in estimates may not test as
statistically significant at the 95%- or even the 90%
confidence level. Significance testing calculations were
conducted at BJS using statistical programs developed
specifically for the NCVS by the U.S. Census Bureau.
These programs consider the complex NCVS sample design
when calculating generalized variance estimates.
Estimates based on 10 or fewer sample cases have high
relative standard errors. Because calculated standard
errors for such estimates may not be accurate, care
should be taken when comparing estimates based on 10 or
fewer cases to other estimates. It is not advisable to
make comparisons between estimates when both are based
on 10 or fewer sample cases. In this report, estimates
based on 10 or fewer sample cases are marked with
Testing trends in weapons violence
Unless stated otherwise when a statement is made
describing differences in estimates between years,
it was tested for significance using a computer
program designed specifically for the NCVS. These
tests determine whether an estimate in one year
differs from that of another, regardless of
intervening estimate variation.
Definitions of crimes:
Aggravated assault - Attack or attempted attack with
a weapon, regardless of whether or not an injury
occurred and attack without a weapon when serious
Rape - Forced sexual intercourse including both
psychological coercion as well as physical force.
Forced sexual intercourse means, vaginal, anal, or
oral penetration by the offender(s). This category
also includes incidents where the penetration is from
a foreign object. Includes rapes of male and female
victims and both heterosexual and homosexual rape.
Attempted rape includes verbal threats of rape.
Robbery - Completed or attempted theft, directly from
a person, of property or cash by force or threat of
force, with or without a weapon, and with or without
Sexual assault - A wide range of victimizations, separate
from rape or attempted rape. These crimes include attacks
or attempted attacks generally involving unwanted sexual
contact between victim and offender. Sexual assaults may
or may not involve force and include such things as
grabbing or fondling. Sexual assault also includes verbal
Simple assault - Attack without a weapon resulting either
in no injury, minor injury (for example, bruises, black
eyes, cuts, scratches, or swelling) or in undetermined
injury requiring less than 2 days of hospitalization. Also
includes attempted assault without a weapon.