Construction of a new secure entryway was among the jobs this past summer at Rosemount Middle School. Nationally, legislators have put more than $950 million into school safety measures this year. (The Minnesota Star Tribune)

Legislators in at least 26 states have poured over $950 million into school safety measures in 2018, but there is very little evidence behind what they funded ("Study Wisconsin's school safety plan," editorial, Dec. 16). Our research has unveiled pathways to school violence prevention that may prove less controversial.

On Dec. 13, a 14-year old student at Dennis Intermediate School in Richmond, Ind., went to school with the intention of committing a mass shooting. After being confronted by police outside the school, he shot and killed himself before anyone else was injured.

Was that a school shooting? It depends on whom you ask.

One problem with preventing mass shootings is that no one can agree on how big the problem is. The Gun Violence Archive reports that there have been 331 mass shootings so far in 2018. Other sources that use the FBI definition (four or more people killed within 24 hours in a public location) put the number at seven.

The definition problem is amplified when we move to school safety issues. The Washington Post recently compiled every incident of gunfire at a primary or secondary school since Columbine in 1999. It identified a total of 220 shootings in 19 years.

The number that meets the FBI definition of a mass school shooting? Six.

For over a year, we have been building a database of all mass shooters since 1966 (using the FBI definition) for a project funded by the National Institute of Justice. In doing so, we reanalyzed the Washington Post's database to identify how many of the 220 school shootings were intended to be mass shootings. That is, the shooter came to school heavily armed and fired indiscriminately at numerous people. Our number? 45.

In the vast majority of cases where a gun was fired in school, the incident was gang-related or was a fight gone south, a targeted act of violence, a domestic or romantic shooting or a suicide. But in 20 percent of cases, the goal of the perpetrator was a mass shooting. That's an average of about 2.5 a year, and the number has been fairly consistent over time.

Who are intended mass school shooters?

The intended mass school shooters were all male. The youngest was 12; the median age was 17. The majority were white, and nearly all were students or former students of the school. In 65 percent of these shootings, no one died.

What types of schools?

Schools that had mass shooting attempts were different from schools with other forms of gun violence in a few ways. They were whiter and wealthier (i.e., fewer students on free lunch). The state with the highest number of attempted mass shootings was California, and most attempts happened on a Monday or Tuesday.

What prevents an intended mass shooting from being a mass shooting?

We also examined what prevented the 39 attempted mass shootings from looking like the six mass shootings with high body counts.

Sometimes the gun jammed or the shooter was a bad aim. Sometimes the police intervened. Sometimes a teacher, student or administrator was able to talk the shooter down. In one case, surrender was negotiated by a janitor.

The presence of a school resource officer didn't make a difference. In fact, the schools with the six deadliest shootings were more likely to have a school resource officer present than the schools where the attempts were foiled. Holding lockdown drills also didn't make a difference.

Warning signs: Our data show that school shooters were more likely to give warning signs before the shootings compared with other mass shooters (87 percent of school mass shooters gave warning signs). A tip to the police is what prevented the Indiana shooting from becoming deadlier. Policies that criminally punish individuals for making threats are not supported by the data. Threats of mass violence are critical for prevention — it's the moment that intervention is crucial. Anonymous reporting systems show promise.

Safe gun storage: By virtue of age, the vast majority of school shooters are unable to buy guns. They beg, borrow and steal guns from adults, many of whom legally own them — most often their family members. One way to prevent students from obtaining these weapons is to safely and securely store them.

Suicide prevention: School mass shootings are almost always murder-suicides. Our research found that school shooters were more likely to be suicidal than other mass shooters (83 percent were suicidal before the shooting). A recent report from the CDC documented epidemic rates of suicide gun deaths last year, the highest number since the Columbine shooting.

Imagine if all teachers and school staff members were as well trained in crisis intervention and suicide prevention as they were in lockdown procedures.

We can do more to create safe schools and prevent mass violence. Our current strategies are not working. The first step is to agree on the scope of the problem. Then we need data to derive, discuss and fund solutions that keep children safe.

Jillian Peterson is a psychologist and professor of criminal justice at Hamline University. James Densley is a sociologist and professor of criminal justice at Metropolitan State University.