EmailPrintCommentRegisterBookmarkRSS What's This

Responding to School Violence with School Violence Experts

The Virginia Model for Student Threat Assessment — from the Public Entity Risk Institute (PERI)

By Dewey G. Cornell, Ph.D.
University of Virginia

This article is brought to you by The Public Entity Risk Institute (PERI). Visit their website: www.riskinstitute.org

Each time there is a highly publicized school shooting, there is widespread concern that school violence is rampant in the United States. For example, after the 1999 Columbine shooting, a Gallup poll found that two-thirds of Americans believed that a similar incident could happen at schools in their community (Saad, 1999). Zero tolerance policies were greatly expanded nationwide so that students were expelled for seemingly minor offenses such as bringing a plastic knife to school, pointing a finger like a gun, or shooting a paper clip with a rubber band (Cornell, 2006).

After several high profile shootings in the fall of 2006, there were renewed recommendations to arm teachers with guns (Associated Press, 2006b) and even a call to issue Kevlar-coated textbooks to students for use as bullet shields (Associated Press, 2006a). A Fort Worth suburban school division went so far as to hire a former military officer to train students to attack and subdue an armed gunman (Dallas Morning News, 2006). Rather than rely on unrealistically heroic measures, it may be more useful to consider less dramatic but more practical prevention approaches that can be implemented well before a gunman appears on school property. Strategies to maintain school safety must be based on a factual assessment of the risk of violent crime and objective evidence of what prevention methods are effective. 

SCHOOL HOMICIDES ARE RARE

Objectively, student-perpetrated homicides are rare events in the nation’s 119,000 schools. There were 103 such cases during the 12 school years from 1992-93 to 2003-04, which means an average of 8.58 per year. Although even one school homicide is too many, an event that affects an average of 8.58 schools out of 119,000 means that the average school can expect a student-perpetrated homicide about once every 13,870 years (119,000 divided by 8.58; Cornell, 2006). Clearly the fear of school homicides as imminent or pervasive events is inaccurate. In contrast, every school must deal with student fights, threats, and bullying on a regular basis. School authorities should not lose perspective on the need for fair and proportionate discipline policies and practices for these frequent problems.

FACTS ABOUT YOUTH VIOLENCE AND SCHOOL SAFETY

The belief that juvenile violence is increasing or that schools are not safe is a serious misconception. According to the Youth Risk Behavior Surveys administered nationally to over ten thousand students each year, there have been declines of 21% in physical fighting and 48% in weapon carrying at school from 1993 to 2003 (Brener, Lowry, Barrios, Simon, & Eaton, 2004). Results from self-report surveys are consistent with other sources of information indicating a downward trend in juvenile violent crime, including violent crime in schools. Although there are limitations to any single source of information about violent crime, the evidence from multiple sources is consistent in indicating an overall decline in school violence from the early 1990’s to recent years. (See charts below.)

Juvenile arrests for homicide in the U.S. According to FBI arrest statistics, juvenile homicide declined over 70% from 1993 to 2005 (FBI, 1993-2005).

Serious violent crime rate per 1,000 students in U.S. schools According to the National Crime Victimization Surveys, serious violent crimes at school (against students 12-18) declined 50% from 1993 to 2003 (DeVoe et al., 2005).

Student-perpetrated homicide cases in U.S. schools Based on media reports tabulated by the National School Safety Center, student-perpetrated homicides at school declined from 42 in 1993 to 13 in 2005.

Students are safer in schools than almost anywhere else. For example, in the year (1999) of the Columbine shooting, 17 students were killed at school, but over 2,500 young people (ages 5 to 19) were murdered outside of school and over 9,700 were killed in accidents (Anderson, 2001). In 2003, students (ages 12-18) experienced twice as much serious violent crime away from school as at school (DeVoe et al., 2005).   

RATIONALE FOR THREAT ASSESSMENT

Although school homicides are rare, the threat of a homicide cannot be dismissed and must be carefully distinguished from more commonplace, less serious threats. Prevention efforts must be informed by an assessment of the factors that precede homicides and how would-be perpetrators can be identified before the shooting starts. Both the FBI and the Secret Service conducted studies of school shootings and found that these students were often victims of bullying who had become angry and depressed, and were influenced by a variety of social, familial, and psychological factors (O’Toole, 2000; Vossekuil, 2002). Unfortunately, these studies concluded that, because these characteristics can be found in so many students, it is not possible to develop a profile or checklist that could be used to pinpoint the small number of truly violent students among them. As a result, both the FBI and Secret Service cautioned schools against a profiling approach.

Nevertheless, the FBI and Secret Service did point out that almost all of these students communicated their intentions to attack through threats and warnings to their peers. Had these threats been reported to authorities and investigated, the shootings might have been prevented. In fact, the FBI identified a number of potential school shootings that were prevented because students reported a threat to authorities that was investigated and determined to be serious. Based on these observations, the FBI and Secret Service both recommended that schools adopt a threat assessment approach to prevent targeted acts of violence (Fein, et al., 2002; O’Toole, 2000). 

What is threat assessment? Threat assessment was developed by the Secret Service to deal with persons who threaten to attack public officials and has evolved into a standard approach to analyze a variety of dangerous situations, such as threats of workplace violence (Fein & Vossekuil, 1998). A threat assessment is conducted when a person (or persons) threatens to commit a violent act or engages in behavior that appears to threaten what is termed “targeted violence.” Threat assessment is a process of evaluating the threat, and the circumstances surrounding the threat, to uncover any facts or evidence that indicate the threat is likely to be carried out. Student threat assessment can be distinguished from profiling in part because the investigation is triggered by some form of student threat behavior rather than some combination of demographic and personal characteristics of the student.

Further, threat assessment is ultimately concerned with whether a student poses a threat, not whether he or she has made a threat (O’Toole, 2000). Any student can make a threat, but relatively few will engage in the planning and preparation necessary to carry out the threat. Threat assessment attempts to identify students who pose a threat, which means that they have the intent and means to carry out the threat. Moreover, threat assessment goes beyond the determination that a student poses a threat to include efforts to prevent the threat from being carried out.

View Threat Assessment Chart (PDF)

THE VIRGINIA MODEL FOR STUDENT THREAT ASSESSMENT

How could the threat assessment approach used in law enforcement be adapted for schools and what would be the results? Our group, the Youth Violence Project of the Curry School of Education at the University of Virginia, sought to answer these questions by developing and field-testing a comprehensive set of threat assessment guidelines (Cornell & Sheras, 2006). We began by convening a work group of educators from two school divisions to help us develop procedures for schools to use in evaluating student threats. The two school divisions contained a socioeconomically and ethnically diverse population of 16,400 students enrolled in four high schools, six middle schools, 22 elementary schools, and three alternative schools. The work group studied how school principals typically handled threats in these 35 schools and attempted to identify common-sense practices and procedures that would be consistent with FBI and Secret Service recommendations.

Threat assessment teams were trained in each school. The teams were led by the principal or assistant principal and typically included a school counselor, a school psychologist, and a school resource officer. (The team composition can be adapted to meet the staffing patterns for different schools, and may include other disciplines as well.) Teams followed a seven-step decision tree. The initial stages of a threat assessment are typically handled by the team leader (principal) and many cases can be readily resolved. In more complex or ambiguous cases, the team leader brings in additional team members. The seven steps will be reviewed briefly here (See figure on page 6; Cornell & Sheras, 2006).

1. At step one, the leader of the threat assessment team interviews the student who made the threat, using a standard set of questions. The principal also interviews the recipient of the threat and any witnesses. The principal is not concerned simply with what the student said or did, but the context in which the threat was made and what the student intended by making the threat.
    
   
2. At step two, the principal must make an important distinction between transient threats, which are easily resolved because they are not serious threats, and substantive threats, which are serious in the sense that they pose a continuing risk or danger to others. Transient threats can be readily identified as expressions of anger or frustration (or perhaps inappropriate attempts at humor) that dissipate quickly when the student reflects on the meaning of what he or she has said. In contrast, substantive threats represent a sustained intent to harm someone beyond the immediate incident. If there is doubt whether a threat is transient or substantive, the threat is regarded as substantive. One way to identify a threat as substantive is to look for certain characteristics derived from the FBI report (O'Toole, 2000) that suggest that the threat is likely to be serious:
   
   
• The threat includes plausible details, such as a specific victim, time, place, and method of assault;
• The threat has been repeated over time or communicated to multiple persons;
• The threat is reported as a plan, or planning has taken place;
• The student has accomplices, or has attempted to recruit accomplices;
• The student has invited an audience of peers to watch the threatened event; and
• There is physical evidence of intent to carry out the threat, such as a weapon or bomb materials.
  
  
3. A transient threat can be resolved quickly at step three without engaging the full team in a comprehensive threat assessment. The principal may require the student to apologize or explain to those affected by threat, or take other action to make amends for the student’s behavior. The principal may also respond with a reprimand or other disciplinary consequence if the behavior was disruptive or violated the school’s discipline code. If a transient threat was sparked by an argument or conflict, the principal can involve other team members in helping to address or resolve the problem. 
   
   
4. If the threat is substantive, the principal skips step three and proceeds to step four. At step four, the substantive threat is determined to be serious or very serious. The distinction between serious and very serious threats is based on the intended severity of injury. A serious threat is a threat to assault, strike, or beat up someone. A very serious threat is a threat to kill, sexually assault, or severely injure someone. A threat involving the use of a weapon is generally considered a threat to severely injure someone.
   
   
5. In the case of a serious substantive threat, the team moves to step five and takes actions to protect potential victims. Protective actions depend on the circumstances of the threat, as well as how soon and where the threat might be carried out. Immediate protective actions include cautioning the student about the consequences of carrying out the threat and contacting the student’s parents. The team also has the responsibility of notifying the intended victim of the threat. 
   
   
6. Very serious threats require the most extensive action by the team. The team skips step five and moves to step six. Again the team takes immediate action to assure that the threat is not carried out, but in addition, the student should be suspended from school, pending a complete assessment of the threat and determination of the most appropriate school placement. The team conducts a more comprehensive safety evaluation that includes both a mental health and law enforcement component. The mental health assessment is conducted by the school psychologist or another suitably trained mental health professional and the law enforcement investigation is conducted by the school resource officer. 
   
   
7. At step seven, the team integrates findings from the safety evaluation into a written safety plan. The safety plan is designed both to protect potential victims and to address the student’s educational needs. At this point, the principal decides whether the student can return to school or should be placed in an alternative setting. If the student is permitted to return to school, the plan describes the conditions that must be met and the procedures in place to monitor the student when he or she returns.

A FIELD TEST OF STUDENT THREAT ASSESSMENT

Over the course of one school year, the 35 schools dealt with 188 student threats. The field test results have been reported in several publications (Cornell et al., 2004a, 2004b; Cornell, 2006) and will be summarized here. The most common threat was a threat to hit or beat up someone (77 cases, 41%). In addition, there were 27 threats to kill, 24 threats to shoot, and 18 threats to cut or stab. There were 32 cases in which the threat was vague or nonspecific (“I’m going to get you”), and 10 miscellaneous threats, such as setting fires or detonating bombs. All types of threats were seen at all school levels, although threats to kill and threats to shoot actually occurred more frequently in elementary school than in middle and high school combined.

The majority (70%) of threats were easily resolved as transient threats. Of the remaining 30% that were substantive threats, 22% (42 cases) were serious substantive threats that involved a threat to fight or assault someone, and 8% (15 cases) were very serious substantive threats to kill or severely injure someone. 

Threats were made by students at all grade levels, from kindergarten through 12th grade. There were 86 elementary school threats, although as might be expected, the vast majority (85%) were determined to be transient and could be resolved through counseling. Elementary school threats often involved rivalries for who would be “best friends” with whom. In some cases, these rivalries generated arguments and threatening statements. In a typical case, a fourth-grade girl wrote a letter making fun of a rival, and then, in reply, received a letter that contained some kind of threat. 
Middle schools (grades 6-8) experienced the highest rate of threats, particularly in the 7th and 8th grades. The 60 middle school threats were almost equally divided between transient (58%) and substantive (42%) cases; many involved an argument or conflict between students that escalated into a threat to fight or assault someone. Many of the threats were precipitated by teasing and bullying behavior; in some cases, the bully was reported for threatening another student, but in other cases, the victim of bullying made a provocative threat in retaliation (e.g., “I’m going to shoot him if he doesn’t leave me alone”).

The 42 high school threats were divided between transient (55%) and substantive (45%) cases. Ninth graders made the most threats; this is not surprising, since ninth graders generally commit more disciplinary violations than any other grade. Threats declined during the high school years, probably because students became more mature and because they became less likely to disclose threats to school authorities.

DISCIPLINARY CONSEQUENCES FOR THREATS

A threat assessment approach gives school authorities flexibility in choosing the disciplinary consequences for students who make threats. Under a zero tolerance policy, many students would have been expelled for making threats to kill or injure someone. Using threat assessment guidelines, only three of the 188 threat cases resulted in expulsion:

• a 6th-grader who picked up a pair of scissors and threatened to stab a classmate;
• an 8th-grader who threatened to shoot a classmate;
• a 9th-grader who threatened to stab another student and was found to have a knife in her locker.

In each case, the student had ten or more disciplinary violations prior to the threat, and the decision to expel was based on a broader consideration of the student’s ability to function in school.

Half (94 cases) of the students who made threats were given a short-term suspension, either an in-school suspension or a suspension outside of school. The modal suspension (32 cases) was 1 day, with a range of one to ten days. 

Twelve students were placed in an alternative educational setting, largely because they had a record of persistent behavior problems. These students had an average of 9.8 disciplinary infractions for the school year prior to their threat.

Only six students were arrested. Three students had made a false bomb threat, two students had assaulted a school staff member, and the final case involved a student found with a knife.

THREAT ASSESSMENT FOLLOW-UP

At the end of the school year, principals were interviewed to obtain follow-up information on each of their cases. The first and most important question was whether the students carried out their threats. In three cases, the principal was not sure whether a student’s threat to hit another student was carried out, but in all other cases, the principals reported that the threat was not carried out.

Principals were also asked about the student’s relationship with the threat recipient after the threat was resolved. They were asked to judge whether the relationship had improved, remained about the same, or worsened. In 62 cases, the principal did not feel sufficiently informed to make a judgment (or there were multiple persons who were threatened), but in the remaining 126 cases, the principals reported that in about one-third of cases the relationship had improved, and in nearly two-thirds (63%) the relationship was about the same; in only six cases (5%) was it worse. 

Finally, principals were asked to assess the student’s overall behavior after the threat. They rated 43% of the students as demonstrating improved behavior during the remainder of the school year, 39% as about the same, and only 18% as worse in their behavior. An independent review of school records found that many of the students continued to have discipline problems. More than half (53%) of the students had a subsequent disciplinary violation of some kind, and 16% had a disciplinary violation that involved violent or aggressive behavior such as hitting someone or getting into a fight.

It seems likely that the process of threat assessment contributed to the positive outcomes found in these schools, but this claim must be made with a caveat: this was a field test and not an experimental comparison between different approaches to student threats. The study was designed to demonstrate the viability of threat assessment as a new and previously untried procedure, but it did not have a comparison group to assess how threats would have been resolved using another method. Currently we are conducting a controlled study that compares schools using a threat assessment model with a comparison group of schools that do not use threat assessment. We also have informal feedback from dozens of schools that have implemented our threat assessment model and found it to be useful. We have provided formal training for over two dozen school divisions in Virginia as well as school divisions in California, Georgia, Indiana, Maryland, Tennessee, and Wisconsin. More information is available at http://youthviolence.edschool.virginia.edu.

THREAT ASSESSMENT AS PART OF A COMPREHENSIVE MODEL

Threat assessment should be considered a component of a comprehensive approach toward maintaining a safe school (Osher, Dwyer, & Jackson, 2004). Threat assessment identifies students who may be in need of additional services as well as more general problems in the school environment, such as bullying, that merit broader attention. Wilson, Lipsey, and Derzon (2003) reviewed 221 studies of school-based interventions for aggressive or disruptive behavior by students, and found that well-implemented demonstration programs are highly effective. 

More broadly, the foundation for a safe school rests on the creation of a caring community where students feel safe and secure (Catalano, et al., 2004). Safety and security derive from two conditions: (1) an orderly, predictable environment where school staff provide consistent, reliable supervision and discipline; and (2) a school climate where students feel connected to the school and supported by their teachers and other school staff. A balance of structure and support is essential, and requires an organized, schoolwide approach that is practiced by all school personnel (Mayer, 1995; Sprague et al., 2002; Sugai et al., 2000;). The good news is that there are effective programs and approaches, and threat assessment can help school authorities to use them effectively and efficiently by identifying student conflicts and problems before they lead to violence.

ABOUT THE AUTHOR

Dewey G. Cornell, Ph. D., is a forensic clinical psychologist and Professor of Education in the Curry School of Education at the University of Virginia. Dr. Cornell is Director of the UVA Youth Violence Project and a faculty associate of the Institute of Law, Psychiatry, and Public Policy. Dr. Cornell has worked with juvenile and adult violent offenders, testified in criminal proceedings and legislative hearings, and consulted on violence prevention efforts. Dr. Cornell has authored more than 100 publications in psychology and education, including studies of juvenile homicide, school safety, bullying, and psychological assessment of psychopathy and violence. His current projects include studies of middle school bullying and guidelines for schools to use in responding to student threats of violence.

References

Anderson, R. N. (2001). Deaths: Leading causes for 1999. National Vital Statistics Reports, 49, 1-88.

Associated Press (2006a). Candidate proposes using textbooks as shields. Retrieved Nov. 12, 2006, from http://www.cnn.com/2006/EDUCATION/10/20/school.shootings.textbooks. ap/index.html

Associated Press (2006b). Wisconsin lawmaker urges arming teachers. Retrieved Nov. 12, 2006, from http://www.usatoday.com/news/nation/2006-10-05-armingteachers_x.htm

Brener, N., Lowry, R., Barrios, L., Simon, T., & Eaton, D. (2004). Violence-Related Behaviors Among High School Students --- United States, 1991—2003. Morbidity and Mortality Weekly Report, 53, 651-55.

Catalano, R.F., Berglund, M.L., Ryan, J.A.M., Lonczak, H.S., & Hawkins, J.D. (2004). Positive youth development in the United States: Research findings on evaluations of positive youth development programs. Annals of the Academy of Political and Social Science, 591, 98-124.

Cornell, D. (2006). School violence: Fears versus facts. Mahwah, NJ: Lawrence Erlbaum.

Cornell, D. & Sheras, P. (2006). Guidelines for responding to student threats of violence. Longmont, CO: Sopris West.

Cornell, D., Sheras, P., Kaplan, S., Levy-Elkon, A., McConville, D. McKnight, L., & Posey, J. (2004a). Guidelines for responding to student threats of violence: Field test of a threat assessment approach. In M. J. Furlong, P. M. Bates, D. C., Smith, & P. M. Kingery, P.M. (Eds.), Appraisal and Prediction of School Violence: Methods, Issues and Contents (pp. 11-36). Hauppauge, NY: Nova Science Publishers.

Cornell, D., Sheras, P. Kaplan, S., McConville, D., Douglass, J., Elkon, A., McKnight, L., Branson, C., & Cole, J. (2004b). Guidelines for student threat assessment: Field-test findings. SchoolPsychology Review, 33, 527-546.

Dallas Morning News (2006). Burleson changes stance on student attack training. http://www.dallasnews.com/sharedcontent/dws/dn/latestnews/stories/102606dnmetburleson.736f1085.html

DeVoe, J. F., Peter, K., Noonan, M., Snyder, T. D., & Baum, K. (2005). Indicators of school crime and safety: 2005 (NCES 2006-001/NCJ 210697). U.S. Departments of Education and Justice. Washington, DC: U.S. Government Printing Office. 

Federal Bureau of Investigation. (1993-2005). Uniform Crime Reports: Crime in the United States. Washington, DC: U.S. Printing Office.

Fein, R.A., & Vossekuil, F. (1998). Protective intelligence and threat assessment investigations: A guide for state and local law enforcement officials. U.S. Secret Service. Washington, D.C.

Fein, R., Vossekuil, B., Pollack, W., Borum, R., Modzeleski, W., & Reddy, M. (2002). Threat assessment in schools: A guide to managing threatening situations and to creating safe school climates. Washington, DC: U.S. Secret Service and Department of Education.

Mayer, G. R. (1995). Preventing antisocial behavior in the schools. Journal of Applied Behavior Analysis, 28, 467-478.

National Center for Education Statistics (NCES). (2005). Indicators of School Crime and Safety.

O'Toole, M. E. (2000). The school shooter: A threat assessment perspective. Quantico, VA: National Center for the Analysis of Violent Crime, Federal Bureau of Investigation.

Osher, D., Dwyer, K., & Jackson, S. (2004). Safe, supportive and successful schools: Step by step. Longmont, CO: Sopris West.

Saad, L. (1999, April 23). Public views Littleton tragedy as sign of deeper problems in country. Retrieved April 2, 2005, from http://www.gallup.com/poll/content/login.aspx?ci=3898

Sprague, J., Walker, H., Golly, A., White, K., Myers, D., & Shannon, T. (2002). Translating research into effective practice: The effects of a universal staff and student intervention on indicators of discipline and school safety. Education & Treatment of Children, 24, 495-511.

Sugai, G., Horner, R.H., Dunlap, G., Hieneman, M., Lewis, T.J., Nelson, C.M., et al. (2000). Applying positive behavior support and functional behavioral assessment in schools. Journal of Positive Behavior Interventions, 2, 131-143.

Vossekuil, B., Fein, R. A., Reddy, M., Borum, R., & Modzeleski, W. (2002). The final report and findings of the Safe School Initiative:  Implications for the prevention of school attacks in the United States. Washington, DC: U.S. Secret Service and U.S. Department of Education. 

Wilson, S. J., Lipsey, M. W., & Derzon, J. H. (2003). The effects of school-based intervention programs on aggressive behavior: A meta-analysis. Journal of Consulting and Clinical Psychology, 71, 136-149.

In January 2006, PERI held a free online symposium to explore strategies for addressing violence in our nation’s schools. The program, entitled "Confronting Violence in Our Schools: Planning, Response, and Recovery," offered practical advice for dealing with school violence — not just the high profile incidents that make headlines but the real issues of school violence that schools and communities face everyday.

The Public Entity Risk Institute (PERI) is a nonprofit research organization that provides education and training resources on topics related to risk management and emergency management. PERI’s website, www.riskinstitute.org, offers free E-Training programs and a Publications, Tools, and Resources library with information on emergency planning and response, disaster recovery, school safety, workers’ compensation, and other key topics in risk management. PERI’s also maintains a national database of public sector liability and workers’ compensation claims data for benchmarking and performance measurement. The Data Exchange is a voluntary program that allows participating organizations to compare their liability and workers’ compensation losses with their peers and learn from other jurisdictions.

The Public Entity Risk Institute (PERI) is a dynamic, forward-thinking organization that serves as a resource to enhance the practice of risk management throughout organizations and communities. Serving public entities, small businesses, and nonprofit organizations, PERI provides relevant and high quality enterprise risk management information, training, data, and data analysis. For more information, please visit www.riskinstitute.org