U.S. Department of Justice
Office of Justice Programs
Bureau of Justice Statistics

****************************************************************
This file is text only without graphics and many of the
tables. A Zip archive of the tables in this report in
spreadsheet format (.csv) and the full report including
tables and graphics in .pdf format are available on BJS
website at: http://www.bjs.gov/index.cfm?ty=pbdetail&iid=5666
****************************************************************

*******************
Special Report
*******************

****************************************
Co-Offending Among Adolescents
in Violent Victimizations, 2004–13
**************************************** 

Barbara A. Oudekerk, Ph.D., and 
Rachel E. Morgan, Ph.D., BJS Statisticians
************************************************

For the 10-year period 2004–13, adolescents ages 12 to
17 committed 50.0 nonfatal violent victimizations per
1,000 adolescents (figure 1). In this report, adolescents
are persons ages 12 to 17, young adults are persons ages 18
to 29, and adults are persons age 30 or older. The rate of
violent victimizations committed by adolescents—or the
adolescent violent victimizing rate—was slightly higher
than the violent victimizing rate for young adults (45.2 per
1,000 young adults). (See Calculating rates using the National
Crime Victimization Survey text box for more information.)
The violent victimizing rate for adolescents was more than
3.5 times higher than the rate for adults age 30 or older
(13.7 victimizations per 1,000 adults).

More nonfatal violent victimizations were committed
by adolescents who acted alone than adolescents who
acted with co-offenders. Adolescents who acted alone
committed 31.8 violent victimizations per 1,000 adolescents.
In comparison, adolescents who acted with at least one
co-offender committed 18.1 violent victimizations per 1,000
adolescents. However, co-offending violent victimizing rates
were higher for adolescents than for young adults (12.9 per
1,000) or adults (1.9 per 1,000).

***********************************************************
************
HIGHLIGHTS
************

This report uses data from the National Crime Victimization
Survey (NCVS) to describe victim, incident, and offender
characteristics of nonfatal violent victimizations committed by
adolescents ages 12 to 17 acting alone or with co-offenders
for the 10-year period 2004 through 2013. Violence includes
simple assault and serious violent victimization (rape or sexual
assault, robbery, and aggravated assault):

* Adolescent offenders who acted alone or with others
committed 50.0 nonfatal violent victimizations per
1,000 adolescents.

* Adolescent offenders committed 22% of all violent
victimizations, while making up 10% of the U.S. population
age 12 or older during this period.

* In violent victimizations committed by adolescents who
acted with at least one other person, co-offenders were most
commonly other adolescents (59%) or young adults (28%).

* More violent victimizations were committed by adolescents
who acted alone (64%) than those who acted with
co-offenders (36%).

* Simple assaults made up a greater percentage of violent
victimizations committed by adolescents acting alone (77%)
or with other adolescents (71%), compared to victimizations
by adolescents acting with young adults (53%).

* A greater percentage of adolescents who acted alone (61%)
or with young adults (70%) used a weapon when engaging
in serious violence than those who co-offended with other
adolescents (50%).

* The rate of serious violent victimizations committed by
adolescent offenders decreased 83% from 1994 to 2013
***********************************************************

This report uses data from the National Crime Victimization
Survey (NCVS) to provide detailed information about
nonfatal violent victimizations committed by adolescents
acting alone or with co-offenders. The NCVS collects data
on victims’ reports of their experiences. The majority of
this report includes only victimizations in which victims
reported the perceived age of their offenders and all
known offenders were age 12 or older (87% of all violent
victimizations from 2004 through 2013).

This report presents aggregate estimates for the period from
2004 through 2013. Trend estimates are based on 2-year
rolling averages centered on the most recent year from 1993
to 2013. For example, estimates reported for 1994 represent
the average estimates for 1993 and 1994. This report refers
to all 2-year estimates by the most recent year. Rolling
averages and aggregating years increases the reliability and
stability of estimates and facilitates comparisons of detailed
victimization characteristics.


***********************************************************
**************************************
Calculating rates using the 
National Crime Victimization Survey
**************************************

The NCVS produces annual victimization, incident, and
prevalence rates for the residential population of the United
States based on a sample from that population. It has also
been used to generate offending rates.***Footnote*Lynch, J. P. & 
Snyder, H. N. (2012). Juvenile crime trends and their 
implications for understanding. Understanding the “Whys” Behind 
Juvenile Crime Trends, 17-49. Available at 
http://whysproject.org/whysFull.pdf.*** This report examines 
victimizing rates, or the rate at which specific groups commit 
victimizations (rather than offenses or incidents), focusing 
specifically on adolescent victimizing rates.

The following examples illustrate the difference between
adolescent victimizing rates and adolescent offending,
victimization, and incident rates. It shows how each rate
would be calculated if three adolescent offenders robbed
three victims--one adult and two adolescents--at gunpoint.

* Victimizing rates are calculated as the number of
victimizations committed by a specific group divided by
the total population of the group. The victimizing rate does
not include the number of offenders who committed the
victimizations. The incident above would contribute three
victimizations to the adolescent victimizing rate because
the adolescents victimized three different people.

* Offending rates are calculated as the number of offenders
from a specific group who are involved in each incident,
summed across all incidents, and divided by the total
population of the group. Offenders may commit multiple
crimes in a given year, and each offender would be counted
separately in each incident. The incident above would
contribute three offenders to the adolescent offending rate
because three adolescents committed the incident.

* Victimization rates are calculated as the number of
victimizations committed against a specific group divided
by the total population of the group. If a crime incident has
more than one victim, each victim is counted separately.
The incident above would contribute two victimizations
to the adolescent victimization rate because two of the
victims were adolescents.

* Incident rates are calculated as the number of incidents
(criminal acts) that occurred within a specific group divided
by that population. An incident is counted only once
regardless of the number of victims. The robbery above
would contribute one incident to the adolescent incident
rate because the two adolescents were robbed in the
same incident.

Both victimizing and offending rates are based on information
about offenders of crime. This report presents victimizing rates
because it includes the number and nature of victimizations
committed by adolescents, and permits estimates of the
amount of victim harm generated by adolescent offenders
who acted alone compared to those who acted with others.
Offending rates would not accurately achieve these goals.

Offending rates would produce duplicate counts of harm
rendered to one victim if the incident was committed by
more than one adolescent offender. For example, three
adolescents committing serious violence against one victim
would be described as “three adolescents committed a serious
violent crime” based on the offending rate, but as “one victim
seriously victimized by adolescents” based on the victimizing
rate. The greater the number of adolescents who committed
victimizations with co-offenders and the greater the number
of co-offenders per victimization, the greater the count of
victim harm. Therefore, offending rates could produce inflated
estimates of victim harm.

Offending rates could also underestimate the amount of
victim harm caused by adolescents. Offending rates count
the number of offenders per incident and not the number of
victims. For example, one adolescent who committed serious
violence against three victims would be described as “one
adolescent committed a serious violent crime” based on the
offending rate but “three persons seriously victimized by an
adolescent” based on the victimizing rate.
***********************************************************

*****************************************************
Adolescents committed 22% of violent victimizations
but made up 10% of the U.S. population
*****************************************************

During 2004–13, adolescents were offenders in 22%
of total violent victimizations known to be committed
by persons ages 12 and older (table 1). In comparison,
adolescents made up 10% of the U.S. population age 12 or
older during this time. Violent victimizations committed
by adolescents involved co-offenders 36% of the time
(table 2). Co-offenders were most often other adolescents
(59%). Another 28% of violent victimizations committed by
co-offending adolescents involved young adult co-offenders.
Victimizations by co-offending adolescents were less
likely to involve co-offending adults (7%) or co-offending
children age 11 or younger (4%). Of the average annual
5,647,900 violent victimizations in which victims reported
offenders’ age, adolescents who acted alone accounted for
14% of victimizations, adolescents who co-offended with
other adolescents accounted for 5% of victimizations, and
adolescents who co-offended with young adults accounted
for 2% of victimizations (table 3).

***********************************************************
*******************************************
Measuring offender age in the National 
Crime Victimization Survey
*******************************************

The NCVS collects information on nonfatal crimes reported
and not reported to the police against persons age 12
or older from a nationally representative sample of U.S.
households. It asks respondents who report a victimization
to indicate whether the crime was committed by a single or
multiple offenders.

When an offender acted alone, the survey asks respondents
to indicate whether they perceive the offender was age 11
or younger; ages 12 to 14, 15 to 17, 18 to 20, or 21 to 29; or
age 30 or older. When multiple offenders committed the
crime, it asks respondents to provide perceived ages of the
oldest and youngest offenders using the same age categories.
Victimizations in which the victim could not estimate the
age of the offenders and victimizations that were committed
solely by offenders age 11 or younger were not included in
this report. (See Methodology for more detail about offenders
age 11 or younger and the limitations inherent in measuring
offender age based on victim reports.)

In this report, offenders were grouped into three age groups:
adolescents ages 12 to 17, young adults ages 18 to 29, and
adults ages 30 or older. These larger age categories make
it possible to reliably examine co-offending subgroups. In
violent victimizations committed by two or more offenders in
2004–13, co-offenders tended to be close in age. For violent
victimizations in which the youngest offender was between
ages 12 to 14, the oldest co-offenders were also ages 12 to 14
in about 46% of the victimizations (table 4). In another 39%
of these victimizations, the oldest co-offenders were ages 15
to 17. Similarly, for violent victimizations in which the age of
the youngest offender was 15 to 17, the oldest co-offender
was most often ages 15 to 17 (42%) or ages 18 to 20 (35%).
The majority (89%) of the youngest co-offenders who
committed violent victimizations with young adults were ages
15 to 17 (table 5).
***********************************************************

Co-offending adolescents were generally the same
race and sex
*************************************************

During 2004–13, adolescents who acted alone to commit
violent victimizations were most often male (72%). When
violent victimizations were committed by co-offending
adolescents, co-offenders were usually the same sex.
Adolescent males were offenders in 62% of violent
victimizations committed by adolescents who acted
with other adolescents (table 6). About 64% of violent
victimizations committed by adolescents who acted with
young adults involved at least one male adolescent offender
who acted with all males.

Violent victimizations committed by a group of adolescents
involved a greater percentage of female offenders (26%)
than those committed by adolescents who co-offended with
young adults (10%). About a quarter (26%) of victimizations
committed by adolescents who co-offended with young
adults involved both male and female offenders.

More than half (59%) of violent victimizations committed
by adolescents who acted alone were committed by white
adolescents, and a quarter (26%) were committed by black
adolescents. Co-offenders were all white or all black in
three-quarters (75%) of violent victimizations committed
by a group of adolescents and in over two-thirds (70%) of
victimizations committed by adolescents who co-offended
with young adults. From 2004 to 2013, the majority (66%)
of victimizations committed by adolescents who acted with
other adolescents involved two or three offenders (table 7).
More than half of violent victimizations committed by
adolescents who acted with young adults involved four or
more offenders (56%), compared to 28% of victimizations
committed by a group of adolescents. When victimizations
were committed by adolescents co-offending with
adolescents or young adults, almost half (42% to 50%)
of victims did not know whether the adolescents were
gang members.

**************************************************
Simple assaults made up the majority of violent
victimizations committed by adolescents
**************************************************

Violent victimizations committed by adolescents were
more likely to be simple assaults than serious violent
crimes (rape or sexual assault, robbery, or aggravated
assault) if adolescents acted alone (77% for simple
assaults compared to 23% for serious violent crimes) or
co-offended with other adolescents (71% compared to
29%) (table 8). In comparison, almost half (47%) of violent
victimizations committed by adolescents who co-offended
with young adults were serious crimes. Most serious
violent victimizations committed by adolescents who acted
alone (14%) or with young adults (27%) were aggravated
assaults. In comparison, most serious violent victimizations
committed by adolescents who co-offended with other
adolescents were robberies (17%).

**************************************************
Most serious violence committed by adolescents 
Did not result in injury
**************************************************

During 2004–13, a weapon was involved in a greater
percentage of serious violent victimizations committed by
adolescents who acted alone (61%) than serious violent
victimizations committed by adolescents who co-offended
with adolescents (50%) (table 9). About two-thirds (70%) of
serious violent victimizations committed by adolescents who
co-offended with young adults involved a weapon.

Although more than half of serious violent victimizations
involved a weapon across these co-offending subgroups, the
majority of adolescent-perpetrated violent victimizations did
not involve weapons (73%), as most adolescent-perpetrated
violence was simple assault. Among all violent victimizations
committed by adolescents, which included serious violence
and simple assault, fewer than 1 in 5 (17%) victimizations
involved a weapon (not shown).

Most serious violent victimizations committed by
adolescents did not result in injury (59% to 70% across
subgroups) (table 10). An injury was involved in a
greater percentage of serious victimizations committed
by adolescents who co-offended with young adults (41%)
than with other adolescents (30%). An injury was involved
in a slightly greater percentage of serious victimizations
committed by adolescents who co-offended with young
adults (41%) than by those who acted alone (32%).

This difference was due to a higher percentage of minor
injuries in serious violence committed by adolescents
who acted with young adults (24%) than alone (17%).
The percentages of serious injuries were not significantly
different for these two co-offending groups. Most injuries
caused by adolescents who co-offended with other
adolescents were bruises, cuts, or other minor injuries (26%
of serious victimizations). Overall, including simple assaults
and serious violence, 24% of total violent victimizations
committed by adolescents resulted in an injury (not shown).
Of these, 21% involved minor injuries and 3% involved
serious injuries (not shown).

**************************************************
Police were notified in 71% of serious violent
victimizations committed by adolescents who acted
with young adults
**************************************************

More than two-thirds (71%) of serious violent victimizations
committed by adolescents who co-offended with young
adults were reported to police during 2004–13 (table 11).

This was higher than in serious violent victimizations
committed by adolescents who co-offended with other
adolescents (51%) or acted alone (46%).

Across all offender ages and co-offending subgroups,
smaller percentages of simple assault than serious violent
victimizations were reported to police. Similar to the
pattern of reporting serious violence to police, a greater
percentage of simple assaults committed by adolescents
who co-offended with young adults were reported to
police (46%) than adolescents who co-offended with other
adolescents (34%) or acted alone (25%). In addition, more
simple assaults committed by adolescents who co-offended
with other adolescents were reported to police, than those
committed by adolescents who acted alone.

**************************************************
A greater percentage of victims were strangers
in serious violence committed by co-offending
adolescents than by adolescents who acted alone
**************************************************

During 2004–13, across all offender age groups, most serious
violent and simple assault victimizations committed by
adolescents were committed against an acquaintance or
stranger. When adolescents committed domestic violence,
they usually offended alone. About 18% of serious violent
victimizations and 8% of simple assaults committed by
adolescents who acted alone were perpetrated against
a current or former intimate partner or family member
(table 12).

A greater percentage of serious victimizations committed by
adolescents who acted alone (48%) or with other adolescents
(42%) were perpetrated against well-known or casual
acquaintances than those committed by adolescents who
co-offended with young adults (21%). Most (67%) serious
victimizations committed by adolescents who co-offended
with young adults were perpetrated against strangers.
Strangers were victims in a greater percentage of serious
violent victimizations committed by groups of adolescents
(56%) than adolescents who acted alone (33%). The pattern
of differences for victim–offender relationships across
co-offending status were generally similar for simple assault
and serious violent victimizations during 2004–13.

**************************************************
Adolescents were generally more likely to victimize
other adolescents than older persons
**************************************************

When adolescents committed serious violent victimizations,
most victims were also adolescents (table 13). Victims
were adolescents in 66% of serious violent victimizations
committed by adolescents who acted alone and in 69% of
those committed by adolescents who co-offended with other
adolescents. In comparison, victims were adolescents in 37%
of serious violence committed by adolescents who acted with
young adults. Victims were more likely to be male in serious
violence committed by adolescents who acted with other
adolescents (64%) or young adults (69%) than by adolescents
who acted alone (51%).

A greater percentage of serious violent victimizations
committed by adolescents who acted alone involved white
victims (63%) than those by adolescents who co-offended
with other adolescents (48%). A greater percentage of
victims were black (27%) or Hispanic (23%) in serious
violence committed by adolescents who co-offended with
other adolescents than those in which adolescents acted
alone (15% each). A greater percentage of victims were
Hispanic in serious violent victimizations committed by
adolescents who co-offended with young adults (22%) than
by adolescents who acted alone (15%).

Most adolescent-perpetrated simple assault victimizations
were committed against other adolescents (table 14).
Victims were adolescents in the majority of simple assaults
committed by adolescents who acted alone (76%) or with
other adolescents (79%). In comparison, victims were
young adults or older in most victimizations committed by
adolescents who acted with young adults (64%). A slightly
greater percentage of victims were male in simple assaults
committed by adolescents who co-offended with young
adults (66%) than by those who acted alone (56%). Victim
race or Hispanic origin distributions were relatively similar
in simple assaults committed by adolescents who acted
alone, with other adolescents, and with young adults.

**************************************************
The adolescent serious violent victimizing rate
decreased 83% from 1994 to 2013
**************************************************

From 1994 to 2013, the adolescent serious violent
victimizing rate decreased from 65.5 to 11.4 serious violent
victimizations per 1,000 adolescents (figure 2). The serious
violent victimizing rate declined similarly from 1994 to 2013
for adolescents who acted alone (30.3 victimizations to 4.8
victimizations per 1,000 adolescents) and adolescents who
acted with co-offenders of any age (35.2 to 6.6 victimizations
per 1,000 adolescents, not shown). From 1994 to 2013, the
serious violent victimizing rate also decreased similarly for
adolescents who acted with other adolescents (16.8 to 3.1
victimizations per 1,000 adolescents) and for adolescents
who acted with young adults (14.6 to 1.6 victimizations per
1,000 adolescents).

The serious violent victimizing rate also decreased 33%
from 2004 to 2013. Adolescents committed 17.1 serious
victimizations per 1,000 adolescents in 2004, compared
to 11.4 serious victimizations per 1,000 adolescents in
2013. The decline from 2004 to 2013 was primarily due to
decreases in serious violence committed by adolescents who
acted alone (9.1 to 4.8 victimizations per 1,000 adolescents)
or with young adults (4.1 to 1.6 victimizations per 1,000
adolescents). The rate of serious violent victimizations
committed by adolescents who co-offended with other
adolescents did not change significantly during this period.

From 1994 to 2013, the adolescent simple assault victimizing
rate decreased 72%, from 139.6 to 38.7 simple assaults per
1,000 adolescents (figure 3). The simple assault victimizing
rate declined from 1994 to 2013 more for adolescents
who acted alone (92.9 to 21.8 simple assaults per 1,000
adolescents) compared to adolescents who acted with
co-offenders of any age (46.7 to 16.9 simple assaults per
1,000 adolescents, not shown). Simple assault victimizing
rates declined similarly from 1994 to 2013 for adolescents
who co-offended with other adolescents (27.7 to 11.8
simple assaults per 1,000 adolescents) and adolescents who
co-offended with young adults (14.9 to 2.4 simple assaults
per 1,000 adolescents).

From 2004 to 2013, the simple assault victimizing rate
decreased 36% among adolescents who acted alone.
Adolescents who acted alone committed 34.1 victimizations
per 1,000 in 2004, which decreased to 21.8 simple assaults
per 1,000 adolescents in 2013. During the same period,
the rate of simple assaults committed by adolescents who
co-offended with other adolescents increased 72%, from
6.9 to 11.8 simple assaults per 1,000 adolescents. The simple
assault victimizing rate among adolescents who acted with
young adults did not change significantly from 2004 to 2013.

****************
Methodology
****************

Survey coverage
******************

The National Crime Victimization Survey (NCVS) is an
annual data collection conducted by the U.S. Census Bureau
for the Bureau of Justice Statistics (BJS). The NCVS is a
self-report survey in which interviewed persons are asked
about the number and characteristics of victimizations they
experienced during the prior 6 months. The NCVS collects
information on nonfatal personal crimes (rape or sexual
assault, robbery, aggravated assault, simple assault, and
personal larceny) and household property crimes (burglary,
motor vehicle theft, and other theft) both reported and not
reported to police. In addition to providing annual level and
change estimates on criminal victimization, the NCVS is
the primary source of information on the nature of criminal
victimization incidents.

Survey respondents provide information about themselves
(e.g., age, sex, race and Hispanic origin, marital status,
education level, and income) and whether they experienced
a victimization. Information is collected for each
victimization incident—about the offender (e.g., age, race
and Hispanic origin, sex, and victim-offender relationship),
characteristics of the crime (including time and place of
occurrence, use of weapons, nature of injury, and economic
consequences), whether the crime was reported to police,
reasons the crime was or was not reported, and experiences
with the criminal justice system.

The NCVS is administered to persons age 12 or older from a
nationally representative sample of households in the United
States. The survey defines a household as a group of persons
who reside at a sampled address. Persons are considered
household members when the sampled address is their
usual place of residence at the time of the interview and
when they have no usual place of residence elsewhere. Once
selected, households remain in the sample for 3 years, and
eligible persons in these households are interviewed every
6 months either in person or over the phone for a total of
seven interviews.

Generally, all first interviews are conducted in-person. New
households rotate into the sample on an ongoing basis to
replace outgoing households that have been in the sample for
a 3-year period. The sample includes persons living in group
quarters, such as dormitories, rooming houses, and religious
group dwellings, and excludes persons living in military
barracks and institutional settings, such as correctional or
hospital facilities, and individuals who are homeless. (For
more information, see Survey Methodology for Criminal
Victimization in the United States, 2008, NCJ 231173, BJS
web, May 2011.)

Weighting adjustments for estimating 
Personal Victimization
****************************************

From 2004 to 2013, an annual average of 144,940 NCVS
interviews were conducted with persons age 12 or older. The
response rate was 87% for eligible persons. Response rates
were lower among adolescents ages 12 to 17 (72%) than
among young adults ages 18 to 29 (81%) and adults age 30
or older (90%). Weighting adjustments in the NCVS have
been found to produce demographic distributions among
adolescents (including sex, race, number of persons in the
household) that are comparable to adolescent demographic
distributions in the American Community Survey.

Victimizations that occurred outside of the United States
were excluded from this report. On average each year
from 2004 to 2013, less than 1% of the unweighted violent
victimizations occurred outside of the United States and
were excluded from the analyses.

Estimates in this report use data from yearly NCVS data
files weighted to produce annual estimates of violent
victimizations for persons age 12 or older living in U.S.
households. Because the NCVS relies on a sample rather
than a census of the entire U.S. population, weights are
designed to inflate sample point estimates to known
population totals and to compensate for survey nonresponse
and other aspects of the sample design.

The victimization weights used in this analysis account
for the number of persons present during an incident and
for repeat victims of series incidents. The weight counts
series incidents as the actual number of incidents reported
by the victim, up to a maximum of 10 incidents. Series
victimizations are victimizations that are similar in type but
occur with such frequency that a victim is unable to recall
each individual event or to describe each event in detail.
Survey procedures allow NCVS interviewers to identify and
classify these similar victimizations as series victimizations
and collect detailed information on only the most recent
incident in the series. In 2013, about 1% of all victimizations
and 4% of all violent victimizations were series incidents.
Weighting series incidents as the number of incidents up
to a maximum of 10 produces more reliable estimates of
crime levels, while the cap at 10 minimizes the effect of
extreme outliers on the rates. Additional information on
the series enumeration is detailed in Methods for Counting
High Frequency Repeat Victimizations in the National Crime
Victimization Survey (NCJ 237308, BJS web, April 2012).

Limitations of using the NCVS to calculate 
Victimizing rates by age of offender
*********************************************

Limitations to calculating violent victimizing rates by age
of the offender using data from the NCVS include missing
respondent data, bias in survey coverage, and differences
in survey response rates by respondent age. This report is
based on victimizations in which the victim provided an
estimate of the offenders’ age. For the period 2004 to 2013,
victims did not provide estimates of offender age in 12%
of the total violent victimizations they reported. Available
data suggest that offenders tend to be close in age to their
victims. In 2004–13, there was no measurable difference in
the percentage of missing data on offender age for adolescent
victims compared to older victims. Data on offender age
were missing for 12% of victimizations committed against
adolescents ages 12 to 17, 12% of victimizations committed
against young adults ages 18 to 29, and 11% of victimizations
committed against adults age 30 or older.

This report is also based on victimizations committed by
at least one person age 12 or older, and excludes about
1% of total violent victimizations against persons age 12
or older that were committed solely by persons age 11 or
younger in 2004–13. These victimizations were excluded
due to insufficient sample sizes and because by design the
NCVS survey coverage does not include children age 11 and
younger. NCVS data likely underestimate the number of
victimizations committed by children age 11 and younger
because the most likely victims of these crimes (i.e., children
age 11 and younger) are excluded from the NCVS sample.
Similarly, NCVS data likely underestimate the number of
adolescent offenders ages 12 to 14 because some of their
victims (children age 11 and younger) are excluded from the
NCVS sample.

The NCVS allows for comparisons between adolescent
offenders ages 12 to 14 and 15 to 17, but small sample sizes
and missing information about co-offenders complicate
these analyses. Grouping offender age into more categories
resulted in insufficient sample sizes to examine many
questions of interest. In victimizations committed by
multiple offenders, it was not always clear whether
adolescents ages 15 to 17 were involved as offenders. For
example, victims might have reported the youngest offender
to be age 12 to 14 and the oldest offender to be age 18 to 20.
In this scenario, it would not be known if one or more of the
offenders were ages 15 to 17. In addition, victims report their
perceptions of the offender’s age. In some victimizations,
particularly those committed by strangers, victims might
inaccurately classify the age of the offender.

Calculating victimizing rates by age of offenders is further
complicated by missing data on the age of all co-offenders.
Respondents are asked to estimate the age of the oldest and
youngest offender. If the youngest offender was reported to
be age 11 or younger and the oldest offender was reported
to be age 18 or older, the involvement of an adolescent age
12 to 17 is unknown. In 2004–13, this made up less than
0.5% of violent victimizations. During the same period,
0.5% of victimizations were reported to be committed
by the youngest offenders (ages 12 to 17) acting with an
oldest offender age 30 or older. Therefore, it was not known
whether a young adult age 18 to 29 was involved.

Offender race and Hispanic origin in the NCVS
************************************************

In 2012, BJS changed the manner in which the NCVS
collects information about the perceived race of offender.
These changes were made to be consistent with the Office
of Management and Budget’s Standards for the Classification
of Federal Data on Race and Ethnicity that specify the
minimum number of categories for statistical reporting
on race.

Information on an offender’s race and Hispanic origin is
collected from the victim, therefore, it is what the victim
perceived it to be. Prior to 2012, the NCVS collected
offender race as white, black or African American, other
race, and don’t know. In 2012, offender race categories were
expanded to include American Indian or Alaska Native; or
Asian, Native Hawaiian, or Other Pacific Islander. In 2012,
the NCVS also began collecting information on whether the
offender was of Hispanic or Latino origin.

Table 6 in this report includes information on the perceived
race of offender. The “other race” category in this table
includes offenders of other races from 2004–11 but also
includes offenders who were American Indian or Alaska
Native; or Asian, Native Hawaiian, or Other Pacific Islander
from 2012–13. Respondents were also asked if the offender
was of Hispanic or Latino origin. However, since it cannot
be determined if an offender was of Hispanic and Latino
origin prior to 2012, this category was excluded from the
analysis for 2012 and 2013. Hispanic or Latino persons were
grouped into their racial category or categories perceived by
the victim. Single offenders of two or more races were also
included in 2012 and 2013.

The “multiple offenders of various races” category in table 6
includes multiple offenders of different races. Prior to 2012,
this included white, black or African American, and other
race. In 2012 and 2013, this included white; black or African
American; American Indian or Alaska Native; and Asian,
Native Hawaiian, or Other Pacific Islander.

Standard error computations
*****************************

When national estimates are derived from a sample, as with
the NCVS, caution must be taken when comparing one
estimate to another or when comparing estimates over time.
Although one estimate may be larger than another, estimates
based on a sample have some degree of sampling error. The
sampling error of an estimate depends on several factors,
including the amount of variation in the responses and the
size of the sample. When the sampling error around an
estimate is taken into account, the estimates that appear to
be different may not be statistically different.

The standard error is one measure of the sampling error
associated with an estimate and can vary from one estimate
to the next. Generally, an estimate with a small standard
error provides a more reliable approximation of the true
value than an estimate with a large standard error. Estimates
with relatively large standard errors are associated with
less precision and reliability and should be interpreted
with caution.

To generate standard errors around numbers and estimates
from the NCVS, the Census Bureau produced generalized
variance function (GVF) parameters for BJS. The GVFs take
into account aspects of the NCVS complex sample design
and represent the curve fitted to a selection of individual
standard errors based on the Jackknife Repeated Replication
technique. The GVF parameters were used to generate
standard errors for each point estimate (such as counts,
percentages, and rates) in this report.

BJS conducted tests to determine whether differences in
estimated numbers and percentages in this report were
statistically significant once sampling error was taken into
account. Using statistical programs developed specifically
for the NCVS, all comparisons in the text were tested
for significance. The Student’s t-statistic was the primary
test procedure, which tests the difference between two
sample estimates.

Data users can use the estimates and the standard errors of
the estimates provided in this report to generate a confidence
interval around the estimate as a measure of the margin of
error. The following example illustrates how standard errors
can be used to generate confidence intervals:

**In 2004 to 2013, according to the NCVS, the rate of
nonfatal violent victimization committed by adolescent
offenders ages 12 to 17 was 50.0 victimizations per 1,000
adolescents (see figure 1). Using the GVFs, the estimate
has a standard error of 2.3 (see appendix table 1). A
confidence interval around the estimate was generated by
multiplying +/- 1.96 (the t-score of a normal, two-tailed
distribution that excludes 2.5% at either end of the
distribution). Therefore, the 95% confidence interval
around the 50.0 estimate from 2004 to 2013 is 50.0 +/-
(2.3 × 1.96) or (45.5 to 54.5). In other words, if different
samples using the same procedure were taken from the
U.S. population from 2004 to 2013, 95% of the time
the rate of nonfatal violent victimization committed by
adolescent offenders ages 12 to 17 would be between 45.5
and 54.5 victimizations per 1,000 adolescents.

In this report, BJS also calculated a coefficient of variation
(CV) for all estimates, representing the ratio of the standard
error to the estimate. CVs provide a measure of reliability
and a means for comparing the precision of estimates across
measures with differing levels or metrics. When the CV
was greater than 50%, or the unweighted sample had 10 or
fewer cases, the estimate was noted with an exclamation
point (! Interpret data with caution. Estimate based on 10 or
fewer sample cases, or the coefficient of variation is greater
than 50%).

**********************************************************
The Bureau of Justice Statistics of the U.S. Department of 
Justice is the principal federal agency responsible for measuring 
crime, criminal victimization, criminal offenders, victims of 
crime, correlates of crime, and the operation of criminal
and civil justice systems at the federal, state, tribal, and 
local levels. BJS collects, analyzes, and disseminates reliable 
and valid statistics on crime and justice systems in the United 
States, supports improvements to state and local criminal
justice information systems, and participates with national and 
international organizations to develop and recommend national 
standards for justice statistics. Jeri M. Mulrow is acting 
director.

This report was written by Barbara A. Oudekerk and Rachel E. 
Morgan, BJS Statisticians. Erika Harrell verified the report.

Lynne McConnell and Jill Thomas edited the report, and Barbara 
Quinn produced the report.

July 2016, NCJ 249756
**********************************************************

*************************************************
Office of Justice Programs
Innovation * Partnerships * Safer Neighborhoods
www.ojp.usdoj.gov
*************************************************

**********************
6/13/2016 1:20pm JER 
**********************