Title: High School Youths, Weapons, and Violence:
A National Survey.
Series: Research in Brief
Author: Joseph F. Sheley and James D. Wright
Published: October 1998
Subject: juvenile violence, crime in schools, gun
violence
18 pages
36,000 bytes

Figures, charts, forms, and tables are not
included in this ASCII plain-text file. To view
this document in its entirety, download the Adobe
Acrobat graphic file available from this Web site
or order a print copy from NCJRS at 800-851-3420.

------------------------------

U.S. Department of Justice
Office of Justice Programs
National Institute of Justice
Jeremy Travis, Director

Research in Brief
October 1998

High School Youths, Weapons, and Violence: A
National Survey

by Joseph F. Sheley and James D. Wright

Recent shootings by students of peers and teachers
in school settings, where such events were
markedly unexpected, have provoked fear and
outrage in America. For many, the "youth-gun
problem" seems to be spreading beyond inner cities
to suburbs and small towns and from "bad boy"
cultures (i.e., those characterized by relatively
high poverty, crime, unemployment, and school
dropout rates) to "good boy" cultures
(characterized by fewer such social ills). Yet,
generalizable and systematic knowledge about
gun-related violence among youths is relatively
scarce. Most national studies to date have asked
only the broadest questions about weapon-related
behavior. Studies using more select samples have
provided more information but, in turn, have been
hampered by questions of generalizability. The
most detailed studies available,[1] for example,
are based on samples of the most serious confined
offenders and recently arrested youths (hardly
populations from which to generalize) and students
from inner-city schools previously identified as
having gun-related problems (rendering
questionable the extent to which the results
pertain to other students, whether from the
suburbs or inner cities).

This Research in Brief examines the extent to
which a national sample of male high school
sophomores and juniors was involved in, or
otherwise affected by, firearm-related activity.
Data were gathered by means of a survey mailed to
high school students. It sought information on
their firearm- and crime-related activities for
the 12 months prior to the survey as well as
social, demographic, and personal information for
each respondent. (The survey contents and goals
are described more fully in "The Tulane University
National Youth Study.")

Despite what is likely a partial "good boy" bias
in the sample (see Methodology), the present study
is the first to pose reasonably indepth questions
to youths from a broad range of social and
geographic environments about exposure to weapons.

Methodology

A random sample of 132 sites was drawn from a
national listing of high schools in the United
States.[2] Of these, 53 (40 percent) consented to
participate in the project. The participating
schools were compared with nonparticipating
schools across numerous variables (e.g., region,
grades offered, size of enrollment, and public or
private status) and characteristics of the cities
and towns in which they were located (e.g., city
size, racial and ethnic population distributions,
age and gender distributions, average educational
attainment level, income characteristics,
employment distribution, percent in poverty, and
crime levels). In all but one instance (a slight
difference in upper age distribution, ultimately
inconsequential), no significant differences were
apparent between those schools that consented to
the survey and those that did not participate.

Completed surveys were received from 734 students
--45 percent of the original sample. The inability
of most schools to conduct followup mailings of
the survey in an effort to increase response rates
and the fact that the sampling procedure was
unable to reach dropouts (who were more likely to
be delinquent) potentially produced a "good boy"
bias in the sample. To check this possibility, the
same survey was administered (with a better than
90-percent response rate) to randomly drawn,
comparative samples of male 10th and 11th graders
on site at three of the schools in our sample.
Onsite respondents differed significantly (always
in the more problematic direction) from
mailed-survey respondents in the following areas:

o School performance. 

o Shooting and beating victimizations off school
grounds.

o Use of knives to threaten others.

o Ownership of automatic or semiautomatic
handguns.

o Gun carrying outside the home.

They did not differ in the following areas:

o Victimizations on school property.

o Victimizations involving knives.

o Arrest history.

o Involvement in theft, burglary, armed robbery,
and assault with a gun or knife.

o Drug use or sales.

o Gang membership.

o Ownership of regular rifles, automatic or
semiautomatic rifles, shotguns and sawed-off
shotguns, and revolvers.

Thus, while a "good boy" bias seems apparent in
the sample, it is not so radical as to render the
sample free of delinquency and related problems.
Indeed, this sample likely more closely
approximates the population of average American
teenagers than have the select samples of most
prior studies that examined youths and their
exposure to firearms.

Respondent characteristics

Respondents in the present study tended to be
between 16 and 18 years of age (97 percent), white
(70 percent), and living with both parents (61
percent), in single-family homes (81 percent),
distributed evenly across the commonly recognized
regions of the country. The respondents' adult
head of the household was likely to have at least
a high school education (81 percent) and to
receive no government assistance (86 percent). The
majority of respondents (90 percent) lived in
towns and cities with fewer than 100,000
residents; 1 in 10 respondents lived in a town
with 2,500 or fewer residents. The average
respondent attended public school (87 percent);
most schools enrolled between 1,000 and 2,500
students (53 percent). Seven of every 10
respondents described their grades as "mostly" Bs
or Cs (73 percent). The average respondent missed
school no more than once monthly (68 percent), had
never been expelled or suspended (67 percent), and
anticipated graduating from high school (94
percent) and immediately enrolling in college (61
percent). 

Fourteen percent of the sample reported having
committed a theft, 8 percent a burglary, and 2
percent an armed robbery during the past 12
months. Five percent reported the use of a "hard"
drug and 3 percent the sale of a "hard" drug.[3]
Only 3 percent of the respondents received
composite "delinquency scores" (i.e., summing
involvement in the behaviors previously noted)
exceeding 2 within a range of 0 to 15 (i.e.,
"never" to "many times" for all offenses).
Finally, 8 percent of the respondents reported a
gang affiliation. In sum, the average respondent
was not seriously involved in criminal activity.

Gun possession and carrying

As the findings in exhibits 1 and 2 indicate,
levels of gun possession and carrying among the
respondents in this study were relatively low, at
least compared with levels reported by more select
samples in prior studies, which focused on males
with higher delinquency rates.[4] Overall, 3 in 10
respondents (29 percent) possessed at least one
firearm (see exhibit 1), and 8 percent possessed
three different types of guns. Moving from an
examination of firearms that are more suited to
hunting and sporting uses (i.e., regular rifles
and shotguns, each possessed by about 2 in 10
respondents), we find that fewer than 1 in 10
respondents possessed a revolver (which could be a
recreational weapon), 1 in 25 an automatic or
semiautomatic handgun (less likely to be a
recreational weapon), and 1 in 50 a sawed-off
shotgun (very unlikely to be a recreational
weapon). 

Gun-carrying behavior was rarer than ownership or
possession (see exhibit 2). Six percent of the
respondents reported carrying a gun outside the
home (including in the car) during the past 12
months--4 percent "now and then" and 2 percent
"most" or "all" of the time. Among gun carriers,
the majority (59 percent) more likely did so in
the car than directly on the person (41 percent).
When a gun was carried outside the home by a
respondent, it most likely was an automatic or
semiautomatic handgun (50 percent) or a revolver
(30 percent).

Urban, suburban, and rural differences

Generally, respondents in smaller communities more
frequently possessed most types of firearms. This
was especially true of rifles (including automatic
or semiautomatic rifles) and shotguns, although
not of sawed-off shotguns. Students from schools
located in rural settings were more likely to
possess revolvers, although automatic or
semiautomatic handgun possession was statistically
no more likely to occur in smaller than in larger
communities. Gun carrying also was more likely to
occur in smaller than in larger communities.

Ease of access to firearms

The findings do little to dispel the common
perception that juveniles can obtain firearms
relatively easily. First, 50 percent of the
present respondents reported that obtaining a gun
would be "little" or "no" trouble if they desired
one; half rated the task as "a lot of trouble" or
"impossible." Second, those who had carried a
handgun outside the home during the past 12 months
were asked where they obtained the handgun.
Forty-eight percent had been given or loaned the
gun by a family member or friend, and 4 percent
reported sneaking the gun from home. Six percent
had stolen or traded something for the gun, while
7 percent had used other, unspecified means of
acquisition. The remaining 35 percent stated that
they had bought the gun (average price, $112),
most commonly from a family member or friend (53
percent of purchasers). Importantly, few
respondents (5 percent) claimed to have asked
someone to purchase a gun (type not specified) for
them from either legal or illegal sources during
the past 12 months.

Recreational gun activity

Respondents were asked to indicate the frequency
(0 through 8 or more times), during the past 12
months, with which they had gone hunting or gone
to a range for target shooting. The items were
moderately correlated (r=0.452; as the hunting
score rose, so also did the target-shooting score)
and were summed with combined scores ranging from
0 to 16. Sixty-one percent of the respondents
received a score of 0, and 17 percent received
scores of 7 through 16. As expected, the
recreational firearm score was negatively, though
only modestly, linked to the respondent's
community setting. As size of city or town
decreased, the recreational firearm score
increased (r=-0.171); as neighborhoods were less
urban in nature, the recreational firearm score
increased (r=-0.240). Not surprisingly, the score
was related to region of the country in which the
respondent resided. Higher scores appeared for the
respondents from the North Central, South Central,
and Mountain States; lower scores characterized
respondents from the New England/Mid-Atlantic,
South Atlantic, and Pacific Coast States.

Recreational use of firearms among the present
sample was associated at statistically significant
levels with possession of every type of firearm of
interest in this study and with carrying firearms
outside the home. Importantly, however, the
recreational firearm score was unrelated to any of
the criminality measures (crime-, drug-, and
gang-related activity) employed in this study, to
measures of status enhancement involving weapons
(e.g., carrying guns to gain respect from peers),
and to indicators of high levels of exposure to
dangerous environments (where gun carrying could
be motivated by a perceived need for
self-protection).

To gain a sense of the potential for distortion in
reports of firearm-related activity when
recreational firearm users are included in
research samples, a restricted sample (N=433) was
created by removing those respondents who
registered a score higher than 0 for the
recreational firearm score (39 percent of the
sample). The amount of gun carrying outside the
home was essentially the same among respondents in
the restricted sample as among the entire sample
but, as indicated in exhibit 3, levels of firearm
possession for the restricted sample were
considerably lower (at least 50 percent less) than
those indicated for the entire sample. The single
exception to this pattern involved the possession
level of automatic or semiautomatic handguns,
which only dropped 25 percent.

It seems, then, that researchers surveying sample
populations in high recreational gun-use areas
(likely more rural in nature) without
simultaneously collecting data on recreational gun
possession, run the risk of inflating the estimate
of nonrecreational gun possession. The likelihood
of overestimating the level of nonrecreational use
or possession of firearms in areas less rural
seems significantly lower. Indeed, once
recreational gun users were taken into account,
most of the significant relationships between the
respondents' environments (e.g., residential and
school populations) and various weapon-related
behaviors and characteristics no longer held.

Illegal activity, status enhancement, and firearms

As noted previously, crime-, drug-, and
gang-related activity characterized only a small
portion of the present sample. Nonetheless, to the
extent that a respondent was engaged in such
criminal activities, his chances of possession of
automatic or semiautomatic handguns, revolvers,
and sawed-off shotguns increased. (Criminal
activity did not enhance possession of regular
rifles, shotguns, and automatic or semiautomatic
rifles.) The likelihood of carrying a gun outside
the home also increased for respondents engaged in
such criminal activities.

Beyond the association between gun-related
activity and criminal behavior, a perceived need
for "respect" is thought by many observers to
motivate juveniles to possess and carry firearms.
However, with few exceptions, firearm-related
activities were not associated with the need for
respect from one's peers among the present
respondents. The exceptions, however, are
important. Possession of automatic or
semiautomatic handguns and carrying a gun outside
the home were related to the need for respect
among peers. To at least some degree, then, status
enhancement was linked to those firearm-related
activities--particularly carrying a handgun
outside the home--that contribute to the increased
use of guns in problematic situations and make the
issue of youths and weapons a serious social
problem.

Guns as protection

This study employed nine items to indicate the
presence or absence of a dangerous social
environment for the respondent. The percentages of
juveniles who answered "yes" to the nine items are
as follows:

o Fear of violence in the neighborhood (48
percent).

o Weapon carrying by friends (14 percent).

o Attendance at social events where shots were
fired (13 percent).

o Firearm- (8 percent) and knife-related (13
percent) threats against the respondent.

o Firearm-related victimization of friends (23
percent) or family members (5 percent).

o Estimated reasonable likelihood of becoming the
victim of a shooting (7 percent) or stabbing (7
percent) by age 25.

Carrying a gun outside the home and possession of
an automatic or semiautomatic handgun were related
to each of the nine dangerous-environment items.
Possession of a sawed-off shotgun was linked to 6
of the 9 items, and possession of a revolver to 5
of the 9. Possession of regular rifles (3 items),
automatic or semiautomatic rifles (2 items), and
regular shotguns (4 items) were less obviously
linked to the dangerous environment in which the
respondent found himself.

The perception that a fairly clear link exists
between dangerous environment and firearm
possession is supported by the responses from
juveniles who had carried guns during the 12
months prior to the survey. Of these respondents,
43 percent cited the perceived need for protection
as the primary reason for bearing arms (see
exhibit 4). Holding a gun for someone was the only
other commonly cited reason for carrying a gun (35
percent). Criminal activity and status enhancement
were relatively less important factors in the
carrying decision.

Policy implications

The findings reviewed in this Research in Brief,
while not derived from a strictly random sample of
high school males, likely come closer to capturing
the firearms experience of the "average" American
juvenile than have many prior studies that focused
on incarcerated delinquents and inner-city youths
or relied on samples from select sites. While this
study may possess a "good boy" bias, it can be
argued reasonably that, relative to the average
incarcerated delinquent, most juveniles are "good
boys." Further, most research indicates that the
ratio of "good" to "bad" boys rises as the focus
moves away from inner-city populations.

The results of this study indicate the need for
policy aimed at reducing the likelihood that
youths (such as those sampled in the present
study) will become involved in firearm possession
and carrying--rather than the more common call for
policy aimed at confronting already well-developed
firearm-related behaviors among youths. What was
found were vastly more students with little or no
experience with weapons and violence than those
with considerable such experience. Even
considering those respondents who subscribed to
the recreational use of weapons, the majority of
students surveyed did not possess weapons, and the
vast majority did not carry them outside the home.
Most of the communities from which respondents
were drawn for this study apparently had not
"crossed the line" into truly unsafe situations.
Yet the line was visible in most. Though
weapon-related violence was perpetrated by few
respondents, many more knew of and were threatened
by it. Further, to the extent that violence was
known, it was shaped by the same variables that
have influenced it in more troubled environments:
crime, drug sales, gangs, and the perceived need
for protection in a hostile world. 

Most schools have adopted the fundamental elements
of persuasion against a culture of violence--some
combination of deterrence (locker searches, for
example) and violence reduction education
(teaching conflict avoidance skills, for example).
(See "Violence Control in Schools.") However,
assigning the responsibility for violence control
and prevention solely or even primarily to the
school curriculum may be dangerous. Though
communities may gain schools that serve as safe
havens and permit education to occur, schools are
less a source of violence than a place where
disputes arising in the neighborhood are acted
upon. To the extent that schools succeed in
pushing violence off school grounds, it probably
will be displaced into the surrounding community.
The issue for communities, then, is how to
dissuade youths from resolving disputes through
violent means, thereby convincing them that
weapons are not necessary for everyday living.
Communities unable to achieve this goal would
turn, understandably, to the criminal justice
system for help.[5] The present findings suggest
that communities in less dire straits should be
exploring policy initiatives that identify and
address the antecedents of weapon-related activity
among their youths.

------------------------------

Issues and Findings

Discussed in this Brief: The firearms experience
of the average youth based on results from an
NIJ-funded 1996 survey of male 10th and 11th
graders from 53 high schools nationwide. 

Key issues: Previous studies have yielded very
little knowledge about firearm-related behavior
applicable to the "average" youth, either because
the study focused on select populations, or
because it asked only the most general of
weapon-related questions. 

Key findings: 

o The number of respondents carrying or in
possession of a gun was relatively low compared
with previous studies: 29 percent possessed at
least one firearm, and 6 percent had carried a gun
outside the home. 

o Gun possession levels were highest for firearms
more suited to hunting and sporting uses: 2 in 10
for regular rifles and shotguns, 1 in 10 for
revolvers, 1 in 25 for automatic or semiautomatic
handguns, and only 1 in 50 for sawed-off shotguns.

o Recreational gun use, more common in rural and
smaller communities, was related to gun-carrying
and possession levels of every type of firearm.

o Gun carrying and possession were more prevalent
in smaller communities. However, once the
recreational gun use variable was held constant,
gun possession among respondents declined; no
effect was seen on gun-carrying levels.

o Fifty percent of the juveniles surveyed felt
that they could obtain firearms relatively easily. 

o Family and friends were the primary sources of
guns (by either giving, lending, or selling the
gun). Few juveniles had asked someone to purchase
a gun for them from legal or illegal sources.

o Although gun-related activity was associated
with crime-, drug-, and gang-related activity,
criminal behavior characterized only a small
number of the respondents. The average juvenile
was not seriously involved in criminal activity. 

o Theft was committed by juveniles (14 percent)
more often than were burglary (8 percent) and
armed robbery (2 percent). 

o A few juveniles reported that they carried
weapons to gain respect from their peers (a form
of status enhancement). While this was rare, the
search for status enhancement increased the
likelihood of carrying weapons.

o For the majority of respondents, the primary
reason for carrying or possessing firearms was
protection, not criminal activity or status
enhancement.

o The majority of school administrators recalled
incidents involving guns on school grounds in the
past 3 years, but only 2 percent considered guns a
serious problem on school grounds.

o Most schools had implemented some
violence-limiting measures. 

Policy implications: To reduce gun-related
violence among youths, policymakers need to focus
on reducing the likelihood that youths will become
involved with firearms; that is, focus on the
antecedents to firearm-related behavior.

Target audience: Law enforcement administrators,
school officials, criminal justice policy
researchers and practitioners, and community
groups that work with youths.

------------------------------

The Tulane University National Youth Study

This project asked participating schools to
facilitate the distribution of a mailed survey to
a sample of 10 percent of their male 10th and 11th
graders in the spring of 1996. Each potential
respondent received a package with the following:

o A letter describing the study and guaranteeing
confidentiality.

o A copy of the survey (containing primarily
forced-choice items).

o A postage-paid return envelope.

o A ticket that, when completed and returned with
the survey, granted the respondent eligibility to
win 1 of 10 cash prizes of $100 to be awarded
through a drawing.

The survey items asked respondents to answer
questions about the following:

o Demographic characteristics.

o Family living situations.

o Educational situations and aspirations.

o Drug, criminal, and gang activities (for the 12
months prior to the survey).

o Crime- and violence-related characteristics of
family and friends.

o Social and recreational activities.

o Exposure to violence generally.

o Personal victimization history.

o Possession of and activities related to firearms
(for the 12 months prior to the survey).

Survey details are available in High School Youth,
Weapons and Violence: A National Survey of
Weapon-Related Concerns, a Final Report to
National Institute of Justice by Joseph F. Sheley
and James D. Wright, May 1998.

------------------------------

Violence Control in Schools

Of the 53 school administrators participating in
this study, 48 responded to a survey concerning
violence at their schools. Differing pictures
emerge depending upon whether we focus on
administrators' estimates of danger on campus or
upon their recollections of weapon-related
incidents on campus. On the one hand, only 2
percent considered guns as at least a somewhat
serious campus problem, and the perception bore no
relation to the urban, suburban, or rural
character of the school. On the other hand, 58
percent recalled incidents involving guns on
school grounds during the past 3 years, and 45
percent reported that at least one of their
students had been shot, on or off school grounds,
during the past 3 years. Importantly, an
administrator's perception of drugs as a campus
problem was highly related to his or her
perception of violence as a problem.

Most schools had instituted some form of response
to the problem of violence. Use of
violence-limiting mechanisms was linked to
perception of drugs as a problem for the school
and, especially, to school dropout rate, but it
generally was not related to the urban, suburban,
or rural setting of the school. Responses to the
"violence problem" among the schools were similar
and not extreme. School administrators reported
the following measures to address violence in
schools:

Measures taken among the majority of the schools

o Automatic suspensions for weapons violations (96
percent). 

o Revised disciplinary codes (81 percent).

o Conflict resolution and multicultural programs
(71 percent).

o Revised dress codes (63 percent).

o Locker searches (55 percent).

Measures taken among fewer schools

o Nonpolice monitors in hallways or on school
grounds (40 percent).

o Photo IDs for staff and students (33 percent). 

o Police on campus (27 percent) or in school
hallways (15 percent).

o Extra police patrols around school property (21
percent).

Measures rarely taken among schools

o Video monitoring of hallways (10 percent).

o ID checks at school entrances (6 percent).

o Metal detectors at entrances (2 percent).

o Mandatory "see-through" book bags (none).

In sum, most of the efforts to prevent violence in
schools here focused on threats of more serious
disciplinary consequences for possession of
weapons and on extensions of traditional school
behavior control (e.g., apparel restrictions or
locker searches). Those schools that moved beyond
such measures tended to rely on external control
agents, usually the police, coming onto campus to
provide security. Few schools had adopted more
extreme measures (e.g., metal detectors and video
surveillance).

------------------------------

Notes

1. Sheley, J., and J. Wright. In the Line of Fire:
Youth, Guns, and Violence in Urban America. New
York: Aldine De Gruyter, 1995; Decker, S., and S.
Pennell. Arrestees and Guns: Monitoring the
Illegal Firearms Market. Report submitted to the
National Institute of Justice, 1996, NCJ 163497.

2. The sample was drawn with sampling
probabilities proportionate to the size of the
10th- and 11th-grade populations enrolled in a
given school. Sampling with probabilities
proportionate to the size of the relevant
population was necessary to avoid the obvious bias
against students in larger high schools.

3. "Hard" drugs refers to heroin, cocaine, and
crack.

4. Sheley, J., and J. Wright. In the Line of Fire:
Youth, Guns, and Violence in Urban America;
Decker, S., and S. Pennell. Arrestees and Guns:
Monitoring the Illegal Firearms Market. 

5. Kennedy, D.,"Pulling Levers: Chronic Offenders,
High-Crime Settings, and a Theory of Prevention,"
Valparaiso University Law Review 31 (1997): 449-
84.

------------------------------

The National Institute of Justice is a component
of the Office of Justice Programs, which also
includes the Bureau of Justice Assistance, the
Bureau of Justice Statistics, the Office of
Juvenile Justice and Delinquency Prevention, and
the Office for Victims of Crime.

------------------------------ 

This study was conducted by Joseph F. Sheley, Dean
of the College of Social Sciences and
Interdisciplinary Studies, California State
University, Sacramento; and James D. Wright,
Favrot Professor of Human Relations, Department of
Sociology, Tulane University. This research was
supported by the National Institute of Justice,
U.S. Department of Justice, through grant number
94-IJ-CX-0033.

------------------------------

NCJ 172857