During the latter part of the 1990s, school violence received widespread attention nationwide and internationally as a result of a small number of traumatic and highly publicized school shootings. In these shootings, which occurred in rural or suburban settings, a student or pair of students carried firearms to school and opened fire, injuring and, in some cases, killing teachers and multiple students. The most infamous of these shootings occurred at Columbine High School in Littleton, Colorado, in 1999, where two students shot and killed fourteen students and one teacher, before turning their guns on themselves and committing suicide. Such events, while relatively rare, were so terrible—the murder of children at their school—that they attracted national and international media attention.

Many people found their perceptions of school violence to be affected by these shootings, which seemed to happen at least once a year from 1996 to 2000. The coverage of these events by the media also enhanced public belief that this type of extreme violence in a school setting was increasing and becoming commonplace. At the same time, various stakeholders— state and local policy makers, law enforcement, school officials—rushed to create and implement polices to predict, prevent, and respond to future incidents of this type of school violence.

The horrific nature of these shootings captured the public’s attentions and stoked their fears, but also, unfortunately, obscured the reality of trends in school violence. Although misbehavior among a small percentage of students is relatively commonplace, major violence is relatively rare in a school setting and, contrary to public belief, decreased during the latter 1990s, and continues to decrease into the first years of the twenty-first century. Furthermore, the schools that dealt with the most widespread violence and disorder were often urban schools rather than suburban or rural schools, where the rare instances of mass violence largely took place. The degree of violence was so great in these shootings, however, that it overshadowed how rarely violence occurred.

During the latter 1990s, as a result of a growing awareness of the problem of school violence, there were also improvements in the recording of such events that revealed a few fundamental findings about school violence in general. First, the traumatic school shootings were rare, isolated events, and not at all reflective of typical school violence. Second, the majority of school violence incidents involve more traditional forms of adolescent violence such as fighting and bullying. Although these are serious events, their results are not even close to the damage and violence that occurred as a result of the school shootings. Moreover, some degree of violence, including bullying and fighting, has been a part of the growing-up experience of adolescents, especially males, for decades. Finally, the majority of incidents of adolescent violence do not happen at school settings or even during school hours. Rather, juveniles most often tended to act violently or become the victims of violence after school, in the afternoon and evening, off school property. Schools, compared to other places that adolescents may go, have become safer and more stable environments in recent years (Devoe et al. 2005).

This article will examine some of the contemporary responses to school violence. First, it will examine the initial response to school shootings. Second, it will examine how violence is recorded. How do stakeholders get their information to measure levels of school violence? At the same time, this entry will report on recent trends in school violence, relying on several different measures, which show serious violence to be relatively rare, with additional decreases in recent years. Finally, this article will discuss the policies that different schools have utilized to prevent and respond to problems with violence. We will examine both responses to serious violence and more commonplace and milder violence, with a summing up of the effectiveness of each approach.

School Shootings during Mid-1990s

During the late 1980s and early 1990s, prior to the highly publicized mass shootings, policy makers created punitive school policies that addressed violence in inner-city schools (Burns and Crawford 1999). The penalties for school violence increased in 1994, when President Clinton signed into law the Gun Free School Safety Act. This law made it a zero tolerance offense to bring a gun or other weapon to school, with the penalty being that the offending student had to be expelled for a year. In response, every state in America created similar gun-free policies.

At the same time, schools also increased their partnerships with police and criminal justice agencies in a way that was for the most part new (Burns and Crawford 1999). While sometimes police had been involved in schools, it was usually in the capacity as an instructor, such as through the use of D.A.R.E., where officers provided education modules on illicit drug use to students. With the changes in policy making, however, police began to engage in more preventive and reactive patrol work in schools, dealing with instances of student behavior, treating such instances as law enforcement issues, and sorting out whether a student had violated a criminal law. While some school systems had experimented with having police in their schools as early as the 1960s, these officers had usually been put into position to foster police and citizen relations. Therefore, the use of criminal justice modalities in the school was largely a new approach for policy making in the schools.

The first high-profile school shootings occurred on February 2, 1996, in Moses Lake, Washington (Washington Post 2000). During the next few years, there were eleven additional high-profile school shootings in other American schools. Of the twelve shootings, three resulted in injuries (no fatalities), three resulted in a single fatality, and six resulted in multiple fatalities. These shootings took place in rural or suburban schools, and often involved Caucasian shooters from middle-class backgrounds, who might have a history of misbehavior in some cases, but nothing to suggest the capability for this level of violence (although some shooters had told other students of their plans).

Due to the media coverage attracted by these shootings, the public began to assume that such events were commonplace and increasing. Burns and Crawford (1999) suggest that the media sensationalized the shootings so that they seemed like commonplace events. The media’s descriptions of the events included phrases such as ”an all-too-familiar story” or ”another in a recent trend” to refer to shootings, as if these were regular events (Donohue, Schiraldi, and Ziedenberg 1998). This was inaccurate. Furthermore, part of the media coverage may have been driven by the fact that this type of violence was so rare and exotic—there was nothing commonplace about it (Donohue et al. 1998).

After the school shootings, Burns and Crawford (1999) suggested that there emerged a moral panic about mass violence at schools. School administrators, teachers, students, and parents began to feel as if violence was inevitable and suddenly perceived certain behaviors or manners of dress as threats to their safety. Wary of future violence, they tried to identify potential shooters, creating profiles of those at risk of committing violence, despite having little information to inform these profiles. Because of the perceived risk of an episode of mass violence, the need to maintain public safety seemed to justify an extreme response.

While previous school violence prevention policies focused on inner-city schools, the reaction to the mass shootings expanded the focus on violence from inner-city schools to suburban schools (Burns and Crawford 1999). This led school administrators to seek ways of enhancing security by creating additional policies and increasing penalties for certain types of behavior. One consequence of the moral panic was the implementation of zero tolerance policies that swept up many students for nonillegal and/or low-level transgressions. According to Burns and Crawford (1999) some schools’ responses included: hiring additional security officers in schools (Sanchez 1998); installing metal detectors in schools (Page 1998) . . . bullet drills in which school children drop and take cover (Tirrell-Wysocki 1998); [and] requiring at least a few school teachers to carry concealed weapons at school (Page 1998). (p. 152)

In the following years, these policies have met resistance as being ineffective and overly punitive. Yet, at the same time, many people believed this was the only response to school violence.

Several students who engaged in low-level offenses ran afoul of these new policies. There has been limited scholarly work that traces the way the moral panic identified such students. Some articles, however, sum up the media accounts from all over the country (Richart, Brooks, and Soler 2003; Skiba 2000). In certain areas, students were prohibited from bringing to school such nonthreatening items as asthma inhalers. Students were arrested and/or expelled for offenses that included bringing to school breath mints, over-the-counter medications, plastic knives in their sack lunches, and pockets knives (Richart et al. 2003; Skiba 2000). Schools have also punished students for pointing their finger at other students in a pretend gun fight and for making comments or jokes about gun violence (Newman 2000; Richart et al. 2003; Skiba 2000).

Many zero tolerance policies also took steps to buttress schools to repel potential attacks. For the zero tolerance polices, schools limited what students could bring to school, ways that students could behave, and words that they could say. Anything that could be perceived as indirectly related to school violence was enough to get a student in serious trouble. In 1999-2000, the target hardening of schools mostly involved limiting the access to school by locking doors, having security fences, school check-ins for visitors to the school, and metal detectors. Each of these policies has been unsuccessful or had limited results (Gottfredson et al. 2004).

Different reports from different states suggest that zero policy programs in schools are unsuccessful and cause harm to those who find themselves ensnared in them. Reports have found that zero tolerance programs are expelling students for drug violations (Potts and Njie 2003), as well as drug and minor violations (Richart et al. 2003). Also there are disparities in who is expelled/punished under programs, with students of color and special education students receiving the most discipline (Potts and Njie 2003; Richart et al. 2003; Michigan Nonprofit Association 2003).

Results are mixed in terms of whether metal detectors have reduced violence. At the end of the 1980s and early 1990s, some sources reported reductions in suspensions for carrying weapons to school and reductions of instances where weapons were brought to school (New York City Board of Education 1991; U.S. General Accounting Office 1995, as cited by Anderson 1998). Other research, however, suggests that metal detectors have little effect on incidents of violence as measured by numbers of victimizations (Schreck, Miller, and Gibson 2003). There have also been issues with how metal detectors are used. Gottfredson and Gottfredson (2001) found that some metal detectors were not plugged in. They also questioned the use of random searches with handheld metal detectors as dangerous: Do you really want to surprise an armed student?

Ways in Which Adolescent Violence Is Recorded

In the last fifteen years changes have occurred in how school violence is recorded. Under the Safe and Drug-Free Schools and Communities Act of 1994, the National Center for Education Statistics (NCES), a federal organization, is mandated to collect information about violence in elementary and secondary schools (OVC Legal Series 2002). The NCES ”is the primary federal entity for collecting, analyzing, and reporting data related to education in the United States and other nations” (DeVoe et al. 2004).

One of the primary sources for getting information about violence in schools comes from the Indicators of School Crime and Safety, an annual report since 1998 that is prepared by the NCES and the Bureau of Justice Statistics. This report contains information about violent and nonviolent crime that happens to students at school or on their way to and from school. The information in the report comes from self-reports of experiences and behaviors from ”students, teachers, principals, and the general population,” and data compiled from several different national surveys.

According to the Indicators of School Crime and Safety (DeVoe et al. 2004), serious violence in schools is a relatively rare event. Between 1992 and 1999, the number of homicides hovered between twenty-eight and thirty-four nationally, and only one to seven suicides, with no increase or decrease (DeVoe et al. 2004, 6). There was a decrease in homicides among students at school for the academic years 1998-1999 and 1999-2000, with thirty-three and fourteen homicides, respectively (DeVoe et al. 2004). The most recent years for which data are available, 2000-2001 and 20012002, show homicides fluctuating between twelve and seventeen incidents at school per year. The more commonplace school violence, student fighting at school, has also declined from 1993 to 2003 (DeVoe et al. 2004, iv). Finally, between 1993 and 2003, there was a decline in the number of students who said they had brought weapons to school from 12% to 6% (DeVoe et al. 2005).

The Indicators of School Crime and Safety also reveals that the school is a safe environment, with more violence occurring off school grounds. During the academic year of 1999-2000, there were 2,124 homicides and 1,992 suicides outside of school property among minors ages five to nineteen; by comparison, there were only twenty-four homicides and eight suicides on school property during that time frame. During the year 2002, students ages twelve to eighteen became victims of serious violence only 88,000 times at school compared to 309,000 outside of school (DeVoe et al. 2004, iv).

There are also other indications that schools are relatively safe places for students. In 2003, 13% of students in grades 9 through 12 said they had been involved in a fight at school, compared to 33% who said they had been in a fight off school grounds (DeVoe et al. 2005). Also, 6% of students in grades 9 through 12 said they had carried a weapon with them at school, compared to 12% who said they had carried a weapon with them anywhere (DeVoe et al. 2005).

In addition to the Indicators of School Crime and Safety reports, many states have passed legislation requiring schools to collect more information about school violence (OVC Legal Series 2002). Many states require schools to maintain and report information on violence, but there are variations across states regarding what types of behavior must be recorded and how much information must be taken. Many states also require schools to report any criminal activity, including violence, to the appropriate police agency. In turn, police agencies in certain states are supposed to reciprocate school reporting by reporting any incidents of juvenile crime that come to their attention to school officials. The OVC Legal Series (2002) identifies the following as continuing issues for law enforcement: ”improvements to statistical reporting,” ”school crime hotlines,” and ”enforcement of reporting laws.”

Finally, it is worth noting that, in total, juvenile violence appears to have dropped according to arrest statistics (Snyder 2005). As of the end of 2003, violent crime arrests of juveniles are lower than they have been since 1987. Between 1994 and 2003, arrests among juveniles dropped 32% for all violent crimes, with drops of 68% for murder, 24% for forcible rape, 43% for robbery, and 26% for aggravated assault. Finally, as of 2003, fewer juveniles are dying as a result of murders, with the numbers at their lowest level since 1984.

Schools Have Different Approaches to Violence

Schools seek to reduce two types of violence: the first being major violent incidents and the second being general violence (bullying, fighting). Reddy et al. (2001) note that students who have engaged in extreme violence are different from those students who engage in lower levels of violence frequently. For extreme situations of violence involving mass shootings, Reddy et al. (2001) note that three types of interventions are commonly used, and describe each as insufficient. These three approaches are (1) ”profiling,” (2) ”guided professional judgment/structured clinical assessment (including the use of warning signs and other checklists),” and (3) ”automated decision making (including the use of actuarial formulas and expert systems).” Profiling involves examining past offenders of mass violence to try to predict who will commit targeted violence (Reddy et al. 2001). There are several problems with using profiling to predict behavior that has yet to occur. First, it is difficult to identify who potentially might commit targeted violence. If the potential offender does not share a specific trait with a past offender, she or he will be wrongly excluded from selection. There is also a lot of room for false positives. There may be several traits that both offenders and non-offenders may share, and therefore it is possible by this commonality to select people who pose no threat. In a report, the U.S. Secret Service described that ”there is no accurate or useful profile of ‘the school shooter’ ” (National Institute of Justice 2002, 12). Also, the report mentions that:

Knowing that an individual shares characteristics, features, or traits with prior school shooters does not advance the appraisal of risk. The use of profiles carries a risk of overidentification, and the vast majority of students who fit any given profile will not actually pose a risk. The use of these stereotypes will fail to identify some students who do, in fact, pose a risk of violence, but who share few characteristics with prior attackers. (p. 13)

A second problem with profiling is the precision of the profile (Reddy et al. 2001). There have been so few events that it is difficult to develop an accurate profile of a potential offender. Also, different profiles include/exclude different events. Information in one profile might not be as inclusive as another profile. Third, it is also difficult to measure the accuracy of the profile, and to determine if the profile has identified the right variables. The profile could have identified the wrong variable (wearing black clothing), which just happens to correlate with the event, rather than to cause the event to occur (carrying a weapon and several rounds of ammunition to school) (Reddy et al. 2001). There are also problems in that the school official may only focus on variables that confirm their suspicions, rather than examining variables that should exclude the individual from the profile. Finally, the use of profiles is unpopular with everyone: school representatives, parents, and students.

Guided professional judgment/structured clinical assessments involve using a professional, such as a psychologist or counselor, to identify someone who could be a possible threat of committing targeted violence (Reddy et al. 2001). There are also flaws with using guided professional judgment/structured clinical assessment. The case rate, which is the number of prior offenders who have committed mass shootings, is too small to be able to use this to identify potentially violent people (Sewell and Mendelsohn 2000). Similar to the problem with creating a profile, there have been so few offenders that it is impossible to construct a theoretical model of a potential offender. Second, there is little preexisting research on this specific type of violence and, at the same time, the literature of general juvenile violence may not be acceptable. Also, the usual assessments and tests of psychologists may not capture a potential for violence of this magnitude.

Automated decision making involves making a database or management system that can identify someone at potential risk of committing targeted violence (Reddy et al. 2001). There are problems with automated decision making because there is little research and no agreement on risk factors. Similar to the public’s growing awareness of serial killers in 1980s and terrorists since 2001, school shooters received a lot of media and pop culture attention, despite being a relatively rare phenomenon. Due to the heavy and protracted attention this rare phenomenon attracts, the public assumes that this is a commonplace event. There is a paucity of research literature about them because students rarely commit mass shootings. Because there are so few, it has been difficult to create databases or management systems, which rely on patterns of behavior from a large sample of subjects. Database and management systems also operate best when they monitor a multitude of different fields, with thresholds in each of these fields set by baselines of the sample they are monitoring. This, then, assumes two things. First, that there is a sufficiently large sample of students generally and, second, that there are enough prior shooters that thresholds can be set from a cross section of their behaviors that will provide a means to recognize and identify future potential shooters. Neither appears to be true in the area of school shootings.

The second set of programs attempts to deal with the problem of violence generally in schools. These programs are interested in preventing the more commonplace types of violence that a larger number of students engage in compared to the rare phenomenon of school shootings. Schools break programs into different groups. Wilson and Lipsey (2005), for example, divide the programs into the following groups:

• Universal programs: ”these programs are delivered in classroom settings to the entire classroom; children are generally not selected individually for treatment but receive treatment simply because they are students in a program classroom. However, schools are frequently selected because they are in low-socioeconomic-status and/ or high-crime neighborhoods. The children in these universal programs may be considered at risk by virtue of their socioeconomic background or neighborhood risk.”

• Selected/indicated programs: ”these programs are delivered to students who are selected especially to receive treatment by virtue of the presence of some risk factor, including disruptiveness, aggressive behavior, activity level, etc. Most of these programs are delivered to the selected children outside of their regular classrooms but are targeted for the selected children.”

• Special schools or classes: ”these programs involve special schools or classrooms that (for the students involved) serve as a usual classroom or school. Children are placed in these special schools or classrooms because of some behavioral or school difficulty that is judged to warrant their placement outside of mainstream classrooms. The programs in this category include special education classrooms for behavior disordered children, alternative high schools, and schools within schools programs.”

• Comprehensive/multimodal programs: ”these programs generally involve multiple modalities and multiple formats, including both classroom-based and pull-out programs. They may also involve programs for parents and capacity building components for school administrators and teachers in addition to the programming provided for the students. The defining characteristic of these programs is that they include multiple treatment elements and formats.” (pp. 10-11)

Schools have also devoted resources to trying to design crime out of schools through changes to the physical school buildings (Gottfredson et al. 2000). Schools have limited access, through locking some doors so that students and visitors have to enter through a specific entrance or entrances. Schools have also required visitors to sign in at the main office of the school. Finally, schools have relied on metal detectors, both stationary detectors that students have to walk through to enter the school, and portable metal detectors that faculty can check specific students with during the day.

Evaluations of Violence Reduction Programs

The programs that have worked for general school violence are universal programs, selected/indicated programs, and compre-hensive/multimodal programs (Wilson and Lipsey 2005). Specifically, these programs reduced behaviors that included ”fighting, name-calling, intimidation, and other negative interpersonal behaviors, especially among higher risk students” (Wilson and Lipsey 2005, 24). The special schools or classes were ineffective at reducing these behaviors.

The impact that these program had on youth violence was related to other variables outside the type of program itself (Wilson and Lipsey 2005). For the successful programs, treatment dose, implementation, and the comprehensiveness of the program were important. Gottfredson et al. (2004) findings also suggest that implementation is important to the success of a program. Further, programs had the greatest impact on higher risk students, because there were greater levels of violence, both in quantity and quality, which could be reduced.

Despite a few meta-analyses and evaluations of programs, there are few examinations of the effectiveness of violence reduction programs. Gottfredson et al. (2004) note that several of the violence strategy programs have not been evaluated, including ”school security practices, architectural arrangements, counseling approaches to problem behavior” (Gott-fredson and Gottfredson 2004, 11).

Next post:

Previous post: