|
Abstract
|
VioLit summary:
OBJECTIVE:
The goal of this article by Garofalo et al. was to examine youth victimization in school-related activities. Methodological innovations with National Crime Survey data and a routine activities approach were used.
METHODOLOGY:
A quasi-experimental method was used for this study. The researchers examined National Crime Survey (NCS) reports for youth 12-17 who had been victimized by offenders who were perceived to be under 18. The offenses examined were the face to face offenses of robbery, aggravated assault, and simple assault; rape and contact larceny occurred too infrequently to be included. The subsample of National Crime Survey crime incidents were chosen through a stratified sampling technique. The strata that were used were type of crime, victim age, offender age, and victim-offender relationship. The Census Bureau sorted the computer files of the NCS Incident Reports from 1982 and 1983. Those strata that exceeded target numbers were randomly subsampled. 429 cases were taken for the sample and were forwarded to NCS for the location of the interviewer narrative and Incident Report. Of 429 cases, 48 were dropped because the hard copies of the reports could not be located, and 8 were excluded because they fell outside sample strata or could not make unique matches with regular incident information on the NCS public use tapes. Subsamples were weighted statistically for analysis. The researchers did a content coding of the narratives for the interviews in terms of the amount and kind of added information they offered about the victim, offender, and incident. Dichotomous coding, 1=present, was used for twelve items. These were drug/alcohol use, gang involvement, victim-offender relationship, offender characteristics, type of items stolen, place of occurrence, self-protective measures, reporting of incident, weapon, victim injury, circumstances surrounding the incident, and intervention by others in the incident. The proportion of the narratives that added additional information ranged from 1% for gangs to 29% for self-protection. Categorical coding was used to reflect the content of the additional information: 1) was prior offending/victimizing behavior involving the victim and offender (s) mentioned, 2) where did the victimization occur, 3) was incident reported to police, 4) was the incident reported to anyone else and to whom, and 5) was the incident interrupted by others and by whom? A quality score was created from the twelve dichotomous variables, and variation in narrative quality was examined across 13 interview, victim, and offender characteristics. There were only small differences found except in age (12 and 13 year old victims had more information) and robbery, street victimization, and victimization by strangers were less likely to contain additional information. The researchers concluded that there was little reason for concern that the use of the narratives would add any systematic biases as the quality was relatively consistent across the variables examined. Descriptive statistics were used to analyze the data.
FINDINGS/DISCUSSION:
The authors took a routine daily activities approach, meaning that it was argued that victimization would take place in the places in which adolescents were usually spending their time. This was the rationale for examining school-related victimizations as they were representative of offenses committed during average adolescent activities. Examination of the sample revealed 850 victimizations with 46% having occurred in school buildings or elsewhere on school property (including school buses). 33% of the victimizations took place in streets or in parks, and 23% occurred in the travel to and from school or waiting for the school bus. A total of 54% were school related; 41% of aggravated assaults, 44% of robberies, and 59% of simple assaults were school related. 75% of the 454 school-related incidents were simple assaults, and the remainder were divided between aggravated assaults (14%) and robberies (12%). The victims and offenders tended to know each other; 79% of the offenders were known to the victims. 23% were well known, 31% were casual acquaintances, and 25% were known by sight only. Out of the victimizations committed by nonstrangers, 28% had additional information that all indicated fellow students, thus, demonstrating a high degree of familiarity. Rather than proposing that schools were highly dangerous places, the authors stated that scuffles, threats, and arguments can fall into NCS categories of more serious activities. Robbery, defined as actual or attempted theft, was demonstrated in 54 cases. 34 of these were completed thefts. Very few of the robberies involved the typical items in adult robberies--cash, purses, wallets, watches, or jewelry. Both successful and attempted robberies involved more personal items such as pencils, books, meal tickets, baseballs, and jackets. Weapons (mostly knives) were used in 25% of the school-related robberies. Of the 65 aggravated assaults, 40% (25) involved the use of a weapon that was included in the "other weapon" category including a diverse set of items including rocks, baseball bats, metal bars, spray paint cans, scissors, screwdrivers and, in one case, a lollypop. Injuries suffered were usually minor--bruises, black eyes, cuts, scratches, and swelling--accounting for 75% of the injuries suffered. In 18% of the aggravated assault narratives and 16% of the simple assault narratives, there was mention that the offender had reacted to a perception of being wronged or insulted; the authors suspected that these figures were underestimates. In general, many school-related victimizations were results of frictions with peers arising from normal daily activities, involved weapons often grabbed at the spur of the moment, and were often characterized by teasing, bullying, and horseplay that got out of hand. The element of presence of capable guardians was also examined. After dividing the sample of school-related victimizations into three place of occurrence categories--inside school building, elsewhere on school property, off school property going to and from school--the authors found a relationship between crime seriousness and the absence of guardians. The two more serious crimes (aggravated assault and robbery) were concentrated in the relatively unsupervised setting of traveling to and from school. Simple assaults were found more often in more supervised settings. Only 15% of the juvenile-to-juvenile robberies and assaults were reported to the police. Among those not reported to police, 39% contained references to other adults being notified. School personnel were found to represent important social control agents with the number being reported to school personnel exceeding those that were reported to police for those narratives that contained additional information. It was generally concluded that school personnel often found themselves in the roles of capable guardians for both deterring victimizations and handling incidents not brought to the attention of the police.
AUTHORS' RECOMMENDATIONS:
The plague of serious school violence reported in the news was not found to have much support, but the frequency at which juveniles do victimize each other was something the authors argued warranted more attention. A deeper understanding of conflict-generating interactions among juveniles for the purpose of devising prevention education was proposed. The potential for use of the systematic narrative descriptions of victimization was restated. Future research, the authors suggested, should take advantage of more open-data collection strategies to better describe dynamics underlying rates of victimization.
EVALUATION:
The National Crime Survey is a recognized and reliable source of data. Its sampling is nation-wide and random. The sample itself is highly generalizable on a large scale. The use of additional data from the survey presents a new dimension to understanding victimization and, potentially, a valuable addition to any NCS analysis. Internal validity is somewhat of a concern, however, though there is no way to correct for it. While the authors carefully show that the distribution for lack of additional information is not patterned, the fact remains that the majority of reports do not have such additions. Furthermore, it is not clear how standardized these narratives are; for example, is there variation in what is defined as "to and from school" and whether it is reported? Further exploration into the narratives as a data source could help establish this internal validity. This study achieves its purpose to be exploratory, but these and other findings using these data need to be subjected to more sophisticated analysis and probability testing to establish that the associations are not spurious and that they are not due to chance. (CSPV Abstract - Copyright © 1992-2007 by the Center for the Study and Prevention of Violence, Institute of Behavioral Science, Regents of the University of Colorado)
KW - 1980s
KW - School Violence
KW - School Victimization
KW - National Crime Survey
KW - Statistical Data
KW - Violence Incidence and Prevalence
KW - Victimization Incidence and Prevalence
KW - Juvenile Violence
KW - Juvenile Victim
KW - Juvenile Offender
KW - Peer Victimization
KW - To and From School
KW - Early Adolescence
KW - Late Adolescence
KW - Junior High School Student
KW - Senior High School Student |