This study examined the effect of exposure to
violence on juveniles. It was specifically concerned with juveniles'
perceptions of violence in schools and communities and how exposure to
violence served as a risk factor for juvenile drug and alcohol use and
participation in other delinquent activities. The study sought to
determine whether exposure to violence increased abuse, dependence,
and regular use of drugs and alcohol, while controlling for
demographic characteristics, family substance use, and peer deviance.
Another objective of the study was to develop a more complete picture
of the context and consequences of violence in schools. In particular,
the study examined the effects of the type of violence witnessed, how
recently the violence was witnessed, the relationship of the offender
and victim to the witness, and whether the witness felt at risk during
the incident.
The data for this study are from the NATIONAL
SURVEY OF ADOLESCENTS IN THE UNITED STATES, 1995 (ICPSR 2833). The
current study drew primarily on the questions that were asked about
respondents' experiences witnessing violence, their own victimization,
peer and family deviance, their own delinquent activities, and drug
and alcohol use. The NATIONAL SURVEY OF ADOLESCENTS IN THE UNITED
STATES, 1995 (ICPSR 2833) consisted of a national probability
telephone sample of 4,023 juveniles between the ages of 12-17 who (1)
were living in United States households with telephones, (2) resided
with a parent or guardian, and (3) could converse in English or
Spanish. All sample selection and interviewing was done by Schulman,
Ronca, and Bucuvalas, Inc. (SRBI), a New York-based survey research
team. All interviews with both parents and adolescents were conducted
using Computer- Assisted Telephone Interviewing (CATI) technology.
After determining that the household contained one or more eligible
adolescents, interviewers asked to speak to a parent or guardian. One
parent or guardian in each household was interviewed briefly to
establish rapport, secure permission to interview the targeted
adolescent, and to ensure the collection of comparative data to
examine potential nonresponse bias from households without adolescent
participation. Parents and guardians were provided the opportunity to
call a toll-free number to confirm the authenticity of the study.
Whenever possible, adolescents were interviewed immediately following
the parent or guardian interviews. Otherwise, appointments were
scheduled when possible or blind callbacks at different times of the
day or days of the week were made. As an incentive for participation,
adolescent participants received a certificate of participation in the
"National Survey of Adolescents" and a check for five dollars as
compensation for their time. From the surveys of parents and
adolescents, the principal investigators created one data file by
attaching the data from the parents to the records of their respective
adolescents.
The NATIONAL SURVEY OF ADOLESCENTS IN THE UNITED STATES,
1995 (ICPSR 2833) consisted of two subsamples, a national probability
household sample of 3,161 adolescents and a probability oversample of
862 adolescents residing in central city areas of the United States,
for a total sample of 4,023. Because the parent or guardian interviews
were conducted prior to the adolescent interviews, the 4,023
participants in the parent sample were also selected from a national
probability sample of households and a probability oversample of
central city households. To construct the initial national probability
sample, a multistage, stratified, area probability, random-digit-
dialing sampling procedure that had four steps was used. First, the
United States was stratified geographically by Census region and a
population-based subsample allocation was developed for each
geographic stratum. In other words, the number of households drawn for
the sample from each geographic stratum was allocated in proportion to
the actual distribution of the population residing within each
stratum, according to the most recent Census estimates. In the second
step, telephone banks within each geographic stratum were
systematically selected utilizing the comprehensive database of
working telephone banks maintained by Schulman, Ronca, and Bucuvalas,
Inc. (SRBI). Third, random-digit-dialing was used to sample telephone
households within the telephone banks selected in the second
stage. Nonworking household (e.g., business) numbers were immediately
replaced by other numbers selected in the same fashion as the initial
numbers. Non-answering numbers were called again four times before
being replaced. In the fourth step, an adult respondent in each
household was screened to determine if there were any adolescents aged
12-17 currently living in the household or if any had lived in the
household at least four months during the previous year. In households
with multiple eligible adolescents, a systematic selection (i.e.,
"most recent birthday" technique) was made to determine which eligible
individual would be designated as the respondent. Construction of the
central city oversample followed the same procedures except for the
initial geographic stratification step. This step was replaced using
the census classification of counties by types of place and specifying
the target population as households located within these urban
counties. Adolescents who were potentially excluded from the study
included those residing in institutional settings, in households
without a parent or guardian (e.g., emancipated minors, married
adolescents living on their own) or in a house without telephones,
those who did not speak English or Spanish, and those whose parents
did not give permission for them to be interviewed.
Adolescents aged 12-17 in the United States.
Individuals.
Data were collected through telephone interviews with
adolescents and their parents or guardians.
survey data
Variables include adolescent respondents' answers
to questions about whether violence and drug abuse were problems in
their schools and communities, what types of violence they had
personally witnessed, where, how recently, and who the victims and
perpetrators were, their alcohol and drug use history, including types
of drugs used, age of first use, frequency of use, and problems
related to use, whether they had been victims of physical or sexual
assault, family background, including whether respondent was subject
to harsh physical punishment and whether anyone in the family had a
problem with drinking or drug use, the delinquent behavior of
respondents and their friends, including destruction of property,
assault, theft, sexual assault, and gang activity, and demographic
information, including age, race, gender, grade in school, and number
of people in household. Parents provided the family's income level.
Other variables include whether the respondent suffered from Post
Traumatic Stress Disorder, a weighting variable based on 1995 census
data, and several derived variables that were not included in the
original dataset, including dichotomous variables related to extent of
drug use, dichotomous variables indicating whether different kinds of
violence were witnessed, and total times different kinds of violence
were witnessed.
Parents in 90.1 percent of eligible households
completed interviews and parents in 78.9 percent of eligible
households gave permission for their adolescents to be interviewed.
Adolescent interviews were completed in 75 percent of eligible
households, 83.2 percent of households with completed parent
interviews, and 95 percent of households with parental permission.
None.