Weapon-Related Victimization in Selected Inner-City High School Samples. MENU TITLE: Weapon-Related Victimization Series: NIJ Published: December 1994 14 pages 29,023 bytes Weapon-Related Victimization in Selected Inner-City High School Samples Joseph F. Sheley, Ph.D. Zina T. McGee, Ph.D. James D. Wright, Ph.D. A Final Summary Report Presented to the National Institute of Justice December 1994 U.S. Department of Justice Office of Justice Programs National Institute of Justice Jeremy Travis Director Lois Mock Project Monitor This project was supported under award number 94- IJ-CX-0033 awarded to Tulane University by the National Institute of Justice, Office of Justice Programs, U.S. Department of Justice. Opinions or points of view expressed in this document are those of the authors and do not necessarily reflect the official position of the U.S. Department of Justice. NCJ 151526 WEAPON-RELATED VICTIMIZATION IN SELECTED INNER-CITY HIGH SCHOOL SAMPLES National victimization survey data indicate that over two million teenagers are the victims of violent crime annually (U.S. Department of Justice, 1992:28). In addition, numerous studies point to the possession of weapons by adolescents as increasingly problematic. This Summary Report explores the issue of weapon-related victimization of inner-city youths attending high schools with histories of violence. It focuses both on levels of such victimization and on characteristics, settings, and activities--including participation in illegal behaviors--that might influence victimization status. BACKGROUND Weapon carrying by youths apparently is neither rare nor isolated. Surveys of inner-city teenage gang members in 3 cities in 1984 and 1985 found that, in a 1-year period, 6 in 10 respondents had carried a weapon with the intention of using it in a fight or had threatened an adult with a weapon; more than half had done so regularly (Fagan 1989). Out of a 1988 sample of Washington, D.C., minority, inner-city 9th and 10th graders, 11 percent had used a weapon to threaten someone (Altschuler and Brounstein 1991). Out of a 1990 nationally representative sample of 11,631 students in grades 9 through 12, 4 percent (21 percent of the black males) reported carrying a gun at least once within the 30 days prior to the survey (U.S. Department of Health and Human Services 1991). Out of a 1991 sample of incarcerated, serious juvenile offenders, 55 percent carried guns all or most of the time prior to incarceration; another 28 percent did so at least occasionally (Sheley and Wright 1993). The problem of youths and weapons apparently has made its way into schools. School administrators increasingly label violence and security major difficulties in their institutions, apparently with good reason. A 1987 survey of high school students in Baltimore, for example, found that almost half of the males had carried a gun to school at least once (Hackett, Sandza, Gibney, et al. 1988). That same year, 3 percent of the males in a survey of 11,000 8th and 10th grade students in 20 States reported bringing a handgun to school during the year preceding the survey (National School Safety Center 1989). A 1990 investigation of 11th grade students in Seattle found that 6 percent had carried a gun to school sometime in the past (Callahan and Rivara 1992). During the 1991-92 school year, 16 percent of the 10th, 11th, and 12th grade students (25 percent of the male students) in an unidentified midwestern urban public school system reported carrying a weapon (not necessarily a gun) to school. Six percent of the same sample engaged in such weapon-carrying more than six times during the school year (Asmussen 1992). FOCUSING ON INNER-CITY YOUTHS Most assaults on adolescents are relatively minor (Whitaker and Bastion 1991; Bastion and Taylor 1991). Perhaps the transport of weapons by youths does not translate to serious injury to youths. However, it may also be that National victimization studies do not focus, at least in sufficient depth, on the weapon-related victimization of that segment of youths popularly thought to experience violence at rates exceeding those of other groups--inner- city juveniles (Altschuler and Brounstein 1991). In this vein, this Summary Report highlights the results of a study of victimization, through firearms, knives, and other weapons, of students attending selected inner-city high schools in four States. It includes a description of victimization levels as well as an examination of important variables linked to victimization. Regarding the latter, special attention is devoted to the extent that involvement in high-risk, illegal behaviors increases the likelihood of violent victimization. The findings derive from responses to surveys completed by 1,591 students (758 males and 833 females) in 10 inner-city public high schools in California, Louisiana, New Jersey, and Illinois. This study was part of a larger investigation of the gun acquisition and possession activities of juveniles that included surveys both of youths in correctional facilities and of students in inner- city schools proximate to those facilities (Sheley and Wright 1993). The sites reflect the instances in which the researchers gained dual entry into both a State's juvenile correction system and at least one adjacent urban school district within reasonably parallel periods. METHOD In all cases, local high school administrators viewed the topic of guns and violence among students as highly politically charged. They consented to the research only with the guarantee that their districts and schools would not be identified in the publication of the research results. Schools selected for study were identified by local school board officials as inner-city schools that had experienced firearms incidents in the recent past and whose students likely encountered gun-related violence (as victims, perpetrators, or bystanders) out of school. No formal evidence exists by which to document these claims. However, interviews with the faculty and students of these schools during the administration of the survey confirmed the administrators' assessments. In one school, the surveyors observed a student remove a gun from his jacket in order to examine it before responding to a particular questionnaire item. In the time since administration of the survey, four of the schools have also experienced violent episodes sufficient to gain national media attention. The survey was introduced to students as a national study of firearms and violence among youths. Participation in the study was voluntary and anonymous. Enrollments in the schools serving as research sites ranged from 900 to 2,100. Principals were asked to grant the research team access to 150 to 200 students in each of the schools entered and, within the practical constraints faced by principals and teachers, to make the sample-- students in grades 9 through 12--as representative of their pupils as possible. At all but two sites this meant access, theoretically, to all students in attendance on the day of the survey. Approximately 95 percent of students addressed by the researchers participated in the study. Percentage of student populations surveyed across schools ranged from 7 to 21 (with a mean of 10 percent; lower percentages were a function of larger-size schools). Responses to the questionnaire items displayed some variation across schools, as expected, but reflected no systematic site-to-site patterns. Site differences that did occur most often reduced to a single site at variance with the others concerning a given item; no one site appeared conspicuously at odds across all items. VALIDITY, COMPLETENESS, AND CONSISTENCY ISSUES Missing data were expected given the nature and length of the survey. Missing cases on items used in the present analysis were contrasted with responding cases controlling for research site, race or ethnicity, and age. Missing and responding cases differed little. As a further check, all analyses reported below were rerun substituting predicted values for missing cases. The results were substantially unchanged. Attempts to establish level of reliability in the present study centered on consistent responses to pairs of items. Eleven such pairs of items were examined. Inconsistent responses averaged only 1.2 percent within a range of .5 to 2.9 percent. To determine how systematic were the inconsistencies, each respondent was scored on the number of inconsistent answers. Respondents received scores between 0 and 11. Only 1 percent of scored above 2; no score exceeded 4. Validity was more difficult to assess since there were no official records against which to compare the self-report data. However, indicative of construct validity, respondents who attributed respect from peers to ownership of a gun also felt that friends would look down on them if they did not carry a gun (r = .549). Levels of use of heroin, crack, and cocaine were associated with extent of commission of property crimes to gain drug money (r ranges between .406 and .458)--a finding consistent with those of previous researchers (Chaiken and Chaiken 1990:212). Validity levels clearly fall within an acceptable range, but see the caveat that follows. CAVEAT It should be stressed that the findings are technically not generalizable to other settings and populations. The four States serving as research sites for this study were not a probability sample of States. Furthermore, to maximize percentages of respondents involved in the behaviors of interest to this study, the study purposely focused on students from especially problematic inner-city schools. Therefore, the 10 high schools and, by virtue of the voluntary nature of participation in the study, the respondents in those institutions serving as research sites were not probability samples of their respective universes. Nonetheless, comparison of respondents' profiles with those known through studies of youths in similar environments indicates that the present sample was not dissimilar to samples from other investigations. RESPONDENT CHARACTERISTICS AND CRIMINAL PROFILES The average respondent's age was 16. The modal educational attainment level was 10th grade. Of those surveyed, 75 percent were African-American, 16 percent Hispanic, 2 percent white, and 7 percent other. All of the respondents were from cities with populations exceeding 250,000. A total of 31 percent of the respondents (43 percent of males and 14 percent of females) reported having been arrested or picked up by the police at least once; 3 percent (6 percent of males, 1 percent of females) had been arrested or picked up "many" times. A total of 15 percent (23 percent of males, 9 percent of females) reported having stolen something worth at least $50, and 5 percent (9 percent of males, 1 percent of females) reported having used a weapon to commit a crime. Much of the recent attention to youths violence has centered on the use and sale of so-called hard drugs, specifically heroin, cocaine, and crack. Any use of hard drugs was reported by only 4 percent (6 percent of males, 2 percent of females). Furthermore, the vast majority of users reported only occasional use. A greater percentage of students (13 percent; 18 percent of males and 9 percent of females) had either dealt drugs themselves or worked for someone who did. The presumed link between gangs and violence is also common in most discussions of crime in the Nation's urban centers. For the present study, gangs are classified into three general types: (1) quasi-gang: a group with whom the respondent identifies but does not define as an organized gang; (2) unstructured gang: a group that is considered an organized gang by the respondent but has fewer than 10 members or has few of the trappings normally associated with gangs (i.e., an "official" name, an "official" leader, regular meetings, designated clothing, and a specified turf); and (3) structured gang: a group that is considered an organized gang by the respondent, has at least 10 members, and has at least 4 of the trappings normally associated with gangs. A total of 22 percent of the students (25 percent of the males, 18 percent of the females) were affiliated with a gang of some kind. Most (55 percent) belonged to quasi-gangs; membership in a structured gang was the least common (16 percent). WEAPON-RELATED VICTIMIZATION Respondents were asked whether they had experienced any of the following while at school or in transit to or from school in the past few years: shot at with a gun, stabbed with a knife, injured with a weapon other than a gun or a knife. Limitation of victimization incidents to those at or associated with travel to and from school necessarily produces a conservative portrait of students' violent experiences. School tends to aggregate youths in a fashion that increases the odds of disputes. However, it also constrains the amount and types of violence that emerge from such conflict. For each of the three victimization items, response categories included: never (0), just once (1), a few times (2), and many times (3). Combining responses to the victimization items permits the calculation of a victimization score for each respondent ranging from 0 (never victimized) to 9 (shot at, stabbed, and otherwise injured many times each). Table 1 displays patterns of victimization by sex. A total of 2 in 10 students had been assault victims. As indicated by the victimization scores, males (30 percent) were twice as likely as females (16 percent) to have been attacked. More than half of those assaulted had experienced multiple victimizations. Type of weapon used in the assaults varied. Students were as likely to have been shot at as injured with a weapon other than a gun or a knife; the odds of having been stabbed were somewhat lower. A higher percentage of males reported shootings. Males were most likely to have been shot at; females most likely were victimized with a weapon other than a gun or knife. DANGEROUS ENVIRONMENTS Attention to the lifestyles or routine activities of victims of violent crime has become a staple of much analysis of victimization patterns. In brief, students of the issue argue that the occurrence of a crime reflects the coincidence of a motivated offender, a suitable target, and the absence of capable guardianship against predation. All else being equal, persons whose routine activities place them in locations more likely frequented by offenders and less well policed than other places stand a better chance of becoming victims of such crimes as robbery and assault. Many of the victim's activities studied by researchers in this area pertain to demographic status; youths and single persons, for example, are at greater risk simply by virtue of lifestyles that take them out of the home and into more trafficked areas, thus increasing their level of exposure as potential targets of violence. Association with members of demographic categories disproportionately populated with predators (e.g., young males) increases the likelihood of victimization (Lauritsen, Laub, and Sampson 1992). Apart from these social category-based activities, certain behavioral choices may put one at higher risk of victimization--accepting rides with strangers, frequenting dangerous bars, walking alone at night in parks, and so forth. That many students in the present study were exposed to dangerous environments is clear. A total of 4 in 10 (39 percent; 37 percent of the males and 42 percent of the females) reported that male relatives carried guns outside their homes. A total of 1 in 3 (35 percent; 42 percent of the males and 28 percent of the females) had friends who carried guns outside the home, and 1 in 4 (23 percent; males and females equally) considered guns easy to get in their neighborhoods. A total of 80 percent of the respondents, males and females equally, reported that other students carried weapons to school; 66 percent of the respondents (again, males and females alike) personally knew someone who had done so. Two-thirds also personally knew someone who had been shot at, stabbed, or otherwise assaulted while in school. Finally, a third of the respondents (38 percent; 35 percent of males and 41 percent of females) agreed or strongly agreed that "there is a lot of violence in this school." HIGH-RISK ILLEGAL ACTIVITIES Nothing in any explication of the routine activities approach specifies that the activities in question must be legal. Indeed, researchers have argued persuasively that a victim's high-risk routine activities as easily can be illegal as legal (Jensen and Brownfield 1986, Lauritsen, Sampson, and Laub 1991, and Sheley, McGee, and Wright 1992). That is, routinely engaging in illegal activities (e.g., drug transactions or robberies) or belonging to groups that routinely engage in or attract violence (sometimes proactively, as do certain youth gangs) would seem to place individuals in exceptionally high-risk categories. As noted earlier, many students in the present sample reported engaging in behaviors that likely increased the risk of violent victimization-- criminal activity, drug use and trafficking, gang activity. Additionally and related, one in four students (males and females roughly equally) reported carrying a weapon while in school. More than 4 in 10 (44 percent; 51 percent of males and 38 percent of females) carried a weapon outside of school. Whether or not weapon-carrying by the students was proactive (related to criminality, for example) or reactive (fear-driven), it can be hypothesized reasonably that the behavior raised the stakes in disputes, leaving all parties at higher risk of injury. WHO IS VICTIMIZED? At least among inner-city students of the kind sampled in this study, sociodemographic characteristics were not highly predictive of violent victimization. Victimization status did not differ significantly among respondents across racial and ethnic lines, age categories, and grade levels. Only sex seemed to influence victimization levels, with males significantly more likely to have experienced a shooting, stabbing, or other weapon-related assault. As the findings in Table 2 suggest, exposure to a dangerous environment significantly raised the risk of weapon-related victimization for respondents of the type studied here. Since the same results pertained for victimization status generally as for each victimization item separately, only relationships regarding the former are reported in the table. Respondents with male relatives who carried guns were more likely than those without such relatives to have been violently victimized. Those whose friends carried guns and those who perceived guns as readily available in the neighborhood also were more likely to report assaults. Students who reported that their peers carried weapons to school and who personally knew other students who had done so were more likely to report having been victimized. Higher percentages of those who personally knew other students who had been shot at, stabbed, or otherwise injured in school reported assaults. Only the perception of one's school as violent was not linked significantly to victimization status, though the direction of this finding mirrors those regarding other measures of environment. Given the findings regarding environment, the data reported in Table 3 concerning higher-risk activities are not surprising. Again, since the same results pertained for victimization status generally as for each victimization item separately, only relationships regarding the former are reported in the table. Criminal activity increased the risk of weapon-related victimization. Those with arrest records, those who had stolen something worth at least $50, and those who had used a weapon to commit a crime were more likely to have been victimized. So also were those who reported carrying weapons in and out of school, using hard drugs, selling drugs, and belonging to a gang. Gang members who belonged to structured gangs were more susceptible to violent victimization. MULTIVARIATE CONSIDERATIONS Few of the variables examined in this study as potentially linked to weapon-related victimization are uncorrelated with the others. It is possible, then, that some of the relationships reported in Tables 2 and 3 actually simply echo others; many may be reduced to a few. To test this possibility, multivariate analysis was conducted to assess the relationship between any given variable and violent victimization independent of the effects of other variables (1). The findings indicated that gender is the only sociodemographic variable related to weapon-related victimization; males are more likely to be victims of violence. The dangerous environment outside of school is related to violent victimization, but the dangerous enviroment inside of school is less obviously related. Independent of the effects of other variables, personal acquaintance with those who carry weapons to school and perceiving one's school as a violent place are not linked to victimization (perhaps indicating rational decisions to maintain a low profile in the dangerous environment). Only personal acquaintance with those who have been victimized in school is related to victimization level (perhaps suggesting the possibility of simultaneous victimization of respondent and friends). High-risk activities, for the most part, increase the likelihood that a student will experience weapon-related victimization. Net of the effects of the other variables, only drug-related activity (drug selling and hard-drug use) is not related to the victimization of students of the type surveyed in this study. IMPLICATIONS Weapon-related victimization clearly is not uncommon among the type of inner-city high school students who were the subjects of this research project--that is, among students in particularly troubled inner-city schools. One in five students (nearly one in three males) among the current respondents had been shot at, stabbed, or otherwise injured with a weapon at or in transit to or from school in the past few years. A total of 1 in 10 had been victimized more than once. Stabbings occurred least often; shootings and injuries from other weapons were roughly equally common. Aside from gender, no sociodemographic characteristic of the current selected sample was linked to victimization level. However, many students in the present sample reported engaging in behaviors that increased the risk of violent victimization-- criminal activity, gang activity, the carrying of weapons. For example, 1 in 4 respondents reported having carried a weapon while in school, and 4 in 10 had carried a weapon outside of school. Perhaps the most striking of the present findings is the apparent level of danger that characterized so many students' social environments. A total of 4 in 10 students reported that male relatives carried guns outside their homes. One in three had friends who carried guns outside the home. One in four considered guns easy to obtain in their neighborhoods. Two of every three respondents personally knew someone who had obtained a gun. Two-thirds also personally knew someone who had carried a weapon to school. At the multivariate level, a dangerous environment outside the school, as opposed to a dangerous environment inside the school, was the better predictor of weapon-related victimization at or during travel to and from school. Judging from the present findings (and recognizing that the present respondents are students in particularly troubled schools), it appears that schools do not generate weapon-related violence as much as they represent the location (exactly or approximately) where violence spawned outside the institution is enacted. That is, inner-city youths do not assume new personae upon passing through school gates. Most weapon-related violence in schools is imported and occurs because the social worlds of some students encourage the use of weapons (students see males in their families carrying guns, for example), because some pupils engage in potentially violent behaviors (criminal activity, for example), or because simply carrying weapons promotes more injurious outcomes of standard juvenile disputes. Schools can do much to prevent violence on their grounds--metal detectors, increased security, and so forth--though the production of high levels of such prevention (e.g., saturated patrols as opposed to sporadic patrols) can be quite costly. It is also clear that differences in administrations, teachers, and physical layouts are linked to differential rates of violence across school sites. Numerous programs are now being introduced by which schools become safe havens from community and home- related problems and through which students are taught basic conflict management skills for use in and out of school. But the source of trouble lies outside the school, and school security efforts are more likely to displace than to reduce violence. Reduction in the levels of violence against students, as opposed to reduction of violence on school grounds, will follow only after the external conditions promoting the violence are addressed. This will require a community-level comprehensive effort. FOOTNOTES 1. The specific multivariate technique employed in this analysis was TOBIT (McDonald and Moffitt 1980), employed because the dependent variable has a high proportion of its cases (79 percent) clustered at a limiting value (zero or "no victimization"). The high degree of intercorrelation among the independent variables in the present study necessitated the analysis of many TOBIT models, alternating highly related variables from model to model. Essentially the same patterns emerged with each model. REFERENCES Altschuler, D. and P. Brounstein. 1991. "Patterns of Drug Use, Drug Trafficking and Other Delinquency Among Inner-City Adolescent Males in Washington, D.C." Criminology 29:589-622. Asmussen, K. 1992. "Weapon Possession in Public High Schools." School Safety, Fall: 28-30. Bastion, L. and B. Taylor. 1991. School Crime: A National Crime Survey Report. Washington, D.C.: U.S. Department of Justice. Callahan, C. and F. Rivara. 1992. "Urban High School Youth and Handguns." Journal of the American Medical Association 267:3038-42. Chaiken, J. M. and M. R. Chaiken. 1990. "Drugs and Predatory Crime." Pp. 203-39 in Drugs and Crime, edited by M. Tonry and J.Q. Wilson. Chicago: University of Chicago Press. Fagan, J. 1989. "The Social Organization of Drug Use and Drug Dealing Among Urban Gangs." Criminology 27:633-69. Hackett, G., R. Sandza, F. Gibney, and R. Gareiss. 1988. "Kids: Deadly Force." Newsweek 11:18-19. Jensen, G. and D. Brownfield. 1986. "Gender, Lifestyles, and Victimization: Beyond Routine Activity Theory." Violence and Victims 1:85-99. Lauritsen, J., J. Laub, and R. Sampson. 1992. "Conventional and Delinquent Activities: Implications for the Prevention of Violent Victimization among Adolescents." Violence and Victims 7:91-108. Lauritsen, J., R. Sampson, and J. Laub. 1991. "The Link Between Offending and Victimization among Adolescents." Criminology 29:265-291. McDonald, J. and R. Moffitt. 1980. "The Uses of Tobit Analysis." Review of Economics and Statistics. 62:318-321. National School Safety Center. 1989. Safe Schools Overview. Malibu, CA: National School Safety Center, Pepperdine University. Sheley, J., Z. McGee, and J. Wright. 1992. "Gun Violence In and Around Inner-City Schools." American Journal of Diseases of Children 146:677- 682. Sheley, J. and J. Wright. 1993a. "Gun Acquisition and Possession in Selected Juvenile Samples." National Institute of Justice, Research in Brief (December). Washington, D.C.: U.S. Department of Justice. U.S. Department of Health and Human Services. 1991. Morbidity and Mortality Weekly Report: Weapon- Carrying Among High School--United States, 1990. (October 11) Washington, D.C.: Department of Health and Human Services, Public Health Service. U.S. Department of Justice. 1992. Criminal Victimization in the United States, 1991. Washington, D.C.: U.S. Department of Justice. Whitaker, C. and D. Bastion. 1991. Teenage Victims: A National Crime Survey Report. Washington, D.C.: U.S. Department of Justice.