U.S. flag

An official website of the United States government, Department of Justice.

Creation of School Shooting Open-Source Database Fuels Understanding

Researchers have created the first open-source catalog of U.S. school shootings to better understand the causes of school-related gun violence and identify intervention points.
Date Published
February 23, 2022

When school shootings occur, they tend to be highly publicized events. Even so, little is known about the overall circumstances that lead to these tragedies because school shootings tend to be studied on an individual basis and in isolation rather than in concert with other shootings that have taken place on school grounds (or in close proximity) that did not result in mass causalities. A collective analysis of these types of events could shed light on larger trends.

To better understand the root causes of these events and identify possible intervention points, NIJ-supported researchers examined two theories of criminology in the context of various types of school shootings, the life-course theory and the situational crime prevention theory.[1] The life-course theory posits that an individual’s bond to society, which evolves over time, is fundamentally related to their tendency toward crime. The situational crime prevention perspective argues that for crime to occur there must first be the opportunity to commit the offense.

These two criminological theories have not been examined extensively in the context of school shootings because of the lack of a large, concentrated dataset focused on these incidents. Therefore, the researchers built The American School Shooting Study (TASSS) to fill the gap – a groundbreaking, national open-source database of all known shootings that resulted in at least one injury on K-12 school grounds between 1990 and 2016, including shootings that took place on school grounds or in close proximity and those known as mass school shootings. Researchers endeavored to provide an evidence-based understanding of the key causes of school shootings by analyzing data on where and when the incidents occur and highlighting key incident and individual-level characteristics.

To benefit law enforcement, school officials, and policy makers, the research team:

  • Built TASSS using open-source data to document the nature of the problem.
  • Analyzed TASSS to provide a comprehensive understanding of the individuals who perpetrated school shootings.
  • Compared fatal and non-fatal attacks to identify possible intervention points.

Researchers also provided carefully curated case studies to capture the life course of those who offend with guns and to gain insight into situational crime prevention related to schools. The ultimate goal was to help law enforcement and school administrators differentiate between the kinds of school shootings that exist to improve policy and response.

Building a Robust TASSS Database from the Ground Up

Building on their past experience creating terrorism and extremist crime open-source databases, the researchers set out to build the TASSS catalog. They collected all publicly available information on each case using strict inclusion criteria. For inclusion in the database, a shooting had to:

  • Have taken place between 1990 and 2016
  • Occur within the continental U.S.
  • Result in a criminal justice response
  • Include a firearm that discharged explosives to propel a projectile.
  • Occur at a K-12 school.
  • Injure or kill at least one person with a bullet wound.

The researchers reviewed over 35 sources and identified almost 1,400 potential shootings.

Of those, 652 shootings fit their criteria. By way of description, they documented that:

  • There were 18 intentional school shootings per year, on average.
  • Mass school shootings are actually rare events compared to all school shootings, with 11 occurring over the 27-year study period.
  • Adult shooters commit more fatal shootings than adolescents.
  • Most of the fatal shootings took place in the 1990s, compared to the 2000s.
  • The South had over three times as many school shootings as the Northeast.
  • A large number of “school shootings” are committed on school grounds, outside the school building by non-students during non-school hours (in other words, are non-school related).
  • Of all the school shootings, fatal shootings are not at an all-time high and do not appear to be steadily increasing.

Looking at life-course theory, the research team identified 252 adolescent school shooters and 102 adult school shooters (including mass school shooters). Upon review of the data, some patterns emerged:

  • Adolescent school shooters were mostly male, with an average age of 16.
  • Most shooters were not employed, and only one quarter of the adult shooters were employed at the time of the attack (which is lower than the general male population).
  • One quarter had psychological issues.
  • One third had criminal records.
  • One third were gang members.
  • One third of adolescents who committed offenses suffered from aggression from their peers, and this was more common with those who committed fatal offenses.
  • Over twice as many fatal shooters than non-fatal shooters had additional family issues.
  • One quarter of school shootings are not intentional shootings against others, but are suicides and accidents.

The case studies reinforce the theory that shootings often result when there is opportunity (i.e., situational crime prevention theory), as identified as easy access to school, gun, or both. The researchers demonstrated that:

  • Consistent with situational crime prevention theory, shootings in multi-story schools with a police officer present were more likely to be non-fatal.
  • Many of those who perpetrated violence were able to enter school with a gun without being accosted, and many were able to access firearms while underage.
  • Many of those who perpetrated violence leaked their intentions – approximately half gave a warning or made threats. Over three quarters of the warnings were not acted on.
  • Contrary to expectations, metal detectors increased the odds of a fatal attack when an adult shooter was involved.

Project Limitations and Future Directions

Importantly, the authors noted that all of the incidents recorded in the TASSS were there because they actually resulted in a school shooting, and not in an attempted shooting. They posited that a useful line of investigation could be to look at “attacked” versus “non-attacked” schools to examine the differential success and use of situational crime prevention interventions like cameras, metal detectors, guards, and officers. It might also be informative to examine foiled plots to see if situational interventions identified the weapon or the shooter prior to the completion of the offense.  

The authors suggested that a one-size-fits-all approach may not work to prevent school shootings. Not all shooters are students, not all victims are students, not all shootings actually take place on school grounds. So how do we prevent intentional shootings? Many that happen to occur on school grounds are simply a manifestation of larger problems on the community level, researchers suggest. These might be best addressed by community partnerships to address larger issues such as gang involvement, drugs, or domestic violence.

The researchers hope to define a more refined typology of those who commit shooting offenses in the future so that they can fashion more targeted interventions. For example, many of those who perpetrated violence leaked their intentions, and over three quarters of the warnings were not acted upon by school officials. Schools can focus the appropriate amount of attention on educating students and staff on the importance of sharing that type of information, no matter how difficult it might prove. Once a refined typology is established, the hope is that possible intervention points will more clearly emerge so that schools can boost their success in preventing future possible school shootings.

About This Article

The work described in this article was supported by NIJ award number 2016-CK-BX-0013, awarded to the Research Foundation of CUNY o/b/o John Jay College of Criminal Justice, CUNY.

This article is based on the grantee report “Understanding the Causes of School Violence Using Open Source Data” (pdf, 84 pages), by J.D. Freilich, S.M. Chermak, N.M. Connell, B. Klein, and E. Greene-Colozzi.

Date Published: February 23, 2022