Student Threat Assessment as a Violence Prevention Strategy

• In Event: School Shootings: Broadening the Context, Refocusing the Response

Abstract

The purpose of this paper is to describe student threat assessment as a violence prevention strategy and to review supporting research evidence from a series of empirical studies conducted in K-12 schools. Concerns about school shootings have stimulated increased school security and use of zero tolerance discipline. The use of exclusionary discipline has increased, resulting in higher rates of school failure, dropout, and criminal justice involvement, especially among Black and Hispanic students. In contrast, threat assessment is a more proactive and flexible violence prevention practice that is used when an individual threatens to commit a violent act or engages in threatening behavior. Threat assessment includes the identification, assessment, and management of threats with the goal of resolving conflicts or problems before they escalate into violence.
A widely-used example of threat assessment is the Virginia Student Threat Assessment Guidelines, which allows school-based teams to follow a decision-tree process to resolve less serious, transient threats quickly while focusing greater attention on more serious, substantive threats. Four controlled studies (summarized in the paper) showed that this approach leads to fewer suspensions and school placement changes, as well as more positive assessments of school climate and safety by school staff and students.
In 2013, Virginia became the first state to require that all public schools have threat assessment teams. In this paper we report recent findings from an NIJ-funded project to examine the use of threat assessment in Virginia, based on a statewide sample of 1,865 cases reported by 785 schools. Logistic regression found that teams were more likely to identify a threat as serious if the student was above the elementary grades and receiving special education services, if the threat involved battery, homicide, or weapon possession, or targeted an administrator. Although few threats were attempted, a threat judged to be serious was about 12 times more likely to be attempted than a threat not judged to be serious. Less than 1% of student threats resulted in a violent act.
In contrast to a zero tolerance approach, teams took actions based on their assessment of the seriousness of the case, reserving the most serious consequences for a small percentage of cases. For example, fewer than half of the students received an out-of-school suspension, approximately 15% received a change in school placement, and less than 1% was expelled. Multilevel logistic regression models found no racial disparities among Black, Hispanic, and White students in disciplinary outcomes. The most consistent predictors of disciplinary consequences were the student’s possession of a weapon and the team classification of the threat as serious. These findings lend support to the use of threat assessment as a less punitive and more equitable approach to student threats of violence. Study limitations and directions for future research will be presented.

Author

• Dewey G. Cornell, University of Virginia