Could a Threat Assessment Have Prevented the Sandy Hook Massacre?

Social scientists try to prevent school shootings by identifying possible perpetrators.


Why? That's what we all want to know. Why did 20-year-old Adam Lanza choose to kill 20 first graders and 6 of their teachers? According to The New York Times, Lanza destroyed the hard drive of his computer and no diary or other notes have so far surfaced. So we may never know his murderous motivations. Could Lanza's rampage have been stopped?

Since mass murders at schools rightly provoke a special kind of horror, lots of social science and criminological research has been focused on identifying possible perpetrators and devising interventions to prevent it. A good review of this research was published in 2002 by the Safe Schools Initiative [PDF], an analysis organized by the Secret Service and the U.S. Department of Education in the wake of the Columbine High School massacre in 1999. That review looked at 37 instances of school shootings involving 41 shooters that took place between 1974 and 2000.  

Who were the attackers? All of them were male, ranging in ages from 11 to 21. Three quarters of them were white and most were living with at least one biological parent. Over 40 percent were doing well in school, receiving As and Bs; 15 percent receiving Bs and Cs; and 22 percent Cs and Ds. Only 5 percent were failing. While 34 percent felt themselves to be "loners," 40 percent socialized with "mainstream" students; while a quarter of them were part of "fringe" high school cliques. In addition, nearly two-thirds of the shooters had never been in trouble or rarely were in trouble. However, 27 percent had been suspended from school and 10 percent had been expelled.

The study confirmed the now-conventional wisdom that school shooters felt bullied, i.e., over 70 percent. Despite the fact that more than 60 percent of the attackers had a history of feeling extremely depressed and more than three-quarters had a history of suicidal thoughts or actions, fewer than one-fifth of them had been diagnosed with mental health issues or behavioral problems.

Media violence—movies, internet, video games—is often blamed for school shootings. In The Hill, Sen. Jay Rockefeller called the Newtown school massacre a "wake up call" for federal action. "While we don't know if such images impacted the killer in Newtown, the issue of violent content is serious and must be addressed," declared the senator. The Safe School Initiatives review reported that about one-quarter of attackers were particularly interested in violent movies and books; one-eighth in violent video games; and more than a third expressed an interest in violence in their own writings.

The report noted, "Incidents of targeted violence at school rarely are sudden, impulsive acts." Almost all of the shooters planned their attacks, and more than half developed their idea for the attack at least a month in advance. Motives for the attacks were often multiple, included revenge (61 percent); to solve a problem (34 percent); desperation (27 percent); and achieve recognition (24 percent). Very disturbingly, in 80 percent of the cases, at least one other person had information that the shooter was planning an attack, usually a sibling, friend, or schoolmate. An adult had prior information about the attack in only two out the 37 cases in the review. The vast majority of shooters did not threaten their targets in advance of their attack. Most attackers had some experience with guns and two-thirds acquired the weapons used in their attacks from their own home or that of a relative. Most attacks ended in less than 15 minutes and law enforcement intervention ended only about a quarter of the incidents.

So how does Adam Lanza match the description of school shooters? Since the Safe School Initiative's definition of "targeted school violence" is limited to "a current student or recent former student," Lanza's attack on the Sandy Hook Elementary School doesn't neatly fit. According to news reports, it is not clear that Lanza ever attended school there. In any case, Lanza was a white male, under age 21, never before in trouble, with an apparent interest in violent video games, and familiar with firearms that he acquired for his attack from his home. So far there is no reported good read on his pre-attack mental state, although he has been described as a shy "loner" who was perhaps somewhere on the autism spectrum. If anyone had any information that might have indicated that he was planning an attack, it would most likely have been his mother who was the first victim of his rampage.

The Safe School Initiative report is quite adamant that there is no reliable "profile" of a school shooter. Of course, it's pretty easy to retrospectively slot shooters into the above categories, but hundreds of thousands (if not millions) of school kids who do not shoot their schoolmates can be pigeonholed into those categories. Given the fact that school shootings are so rare and that past shooters differ considerably in their personalities, and their demographic and social characteristics, constructing prospective profiles able to predict which students are a risk for committing school massacres is impossible. "There is no accurate or useful profile of students who engaged in targeted violence," conclude the researchers.

The researchers at the Safe School Initiative suggest that one possibly effective way to prevent school shootings is a process of threat assessment [PDF] that looks at behaviors rather than an individual student's "traits." Threat assessment asks if a particular student is exhibiting behaviors that suggest preparations for an attack; how fast is he moving toward an attack; and where might intervention be possible? 

The Safe School Initiative recommends that schools set up threat assessment teams with a designated person as the widely known central contact for any information regarding possible threats. What kind of information? Perhaps a student writes a story for English class about a character who shoots other students; or a student overhears a conversation in which another student vows to "get even for good." Recall that in 80 percent of the cases reviewed by the Safe School Initiative, at least one other person had information about the upcoming attack. However, many schools have a "no snitching" culture among students that is a barrier to getting relevant information to teachers, counselors, and administrators. The report advises that schools establish cultures of mutual respect among students and faculty. Well, yes.

One good example of how this might be done is a school district that regularly asks faculty members to identify those students with whom they have the closest relationships by putting a star next to their posted names at staff meetings. The staff then concentrates on establishing relationships with those students with fewest stars.

Considering that 70 percent of school shooters felt bullied, steps to reduce bullying might help prevent future school massacres. Fourteen percent of students, ages 12 through 18, reported being bullied during school in 2001, rising to 28 percent in 2005 and 32 percent in 2007, but falling to 28 percent in 2009. Forty-nine states now have anti-bullying statutes. One might hope that the increase in bullying statistics stems from recent anti-bullying campaigns making students more comfortable about reporting it rather than an absolute increase. Nevertheless, one survey found that only 36 percent of cases of bullying are reported. "No snitching" to teachers is apparently still the rule among teenagers.

Ultimately, the most important aspect of the threat assessment approach is that it be based on the facts of a particular case and on a specific student's behaviors, not "traits" that allegedly characterize would-be school shooters. Does it work? Non-events are hard to document. However, there is at least one case in which researchers argue that a student targeted by a threat assessment methodology in a California vocational school was ill-treated. The student made a vague threat to kill the school district superintendent. He was removed from school and placed in an alternative school for kids with behavioral problems. Violating the rule that such assessments should be based on facts and behaviors, there was no evidence of any planning on the student's part. In addition, the assessment targeted (profiled) the alienated clique of skater kids that he hung out with.

Would the threat assessment approach have identified Adam Lanza as a risk? Given the information currently available, the answer is no. So far there is no evidence Lanza told anyone of his plans. As for bullying, he may well have felt bullied during his relatively short stint in public schools, but by most accounts teachers and administrators had identified him as a likely victim and took steps to prevent it. Did anyone know if Lanza had an unusual interest in targeted violence? Feelings of hopelessness? Show evidence of planning? Worry other people about his potential for harm? Given his isolation from the community, it is unlikely that any sort of threat assessment methodology would have identified him as a potential mass murderer of first graders.

I agree with President Barack Obama that "meaningful action" should be taken prevent future schoolhouse carnage. Meaningful action in this case would be fashioning schools that respect all of their students; protect them from bullying; foster enough mutual trust to curtail "no snitch" teenage culture; and offer students proactive counseling on how to handle their emotional challenges. But that's a whole lot harder than grandstanding about banning assault rifles or violent video games.