74 Interview: Criminologist Nadine Connell on Why She’s Building a 30-Year Database of School Shootings — and What Hidden Lessons May Be Found in the Stats

See previous 74 interviews, including former U.S. Department of Education secretary Arne Duncan, Teacher of the Year finalist Nate Bowling, and two Texas middle schoolers who seek an end to gun violence. The full archive is right here.

Quickly after a gunman opened fire on students and teachers at Marjory Stoneman Douglas High School in Parkland, Florida, the divisive debate over American gun violence — historically focused on school shootings — re-emerged in full force.

Though school shootings have fueled the gun control debate for more than a century, reliable data on the frequency of school shootings and the factors that motivate the gunmen remain hard to come by.

That’s where Nadine Connell, an associate professor of criminology and director of the Center for Crime and Justice Studies at the University of Texas at Dallas, comes in. Connell is building a database of all school shootings in the U.S. since 1990, information she hopes will help law enforcement and school leaders better understand the causes of school violence, and how to prevent it.

The work comes as students and their advocates nationwide demand new policies they say could help keep students safe. Students across the country recently participated in a daylong school walkout to push for new gun control legislation. But, as Connell has found, school shootings are extremely rare. Although the fear of mass school shootings often drives policy, she said, those efforts aren’t always the most productive.

With the help of a federal National Institute of Justice grant, Connell has teamed up with several University of Texas at Dallas students, along with researchers at the University of Maryland at College Park, Michigan State University, and the John Jay College of Criminal Justice at The City University of New York.

As the nation reels from yet another school massacre — this time in Santa Fe, Texas — The 74 spoke with Connell about what motivated the researchers to study school shootings in-depth, and how she hopes their work will influence the debate over school safety. This conversation was edited for length and clarity.

The 74: What prompted you to track school shootings in this way, and what do you hope to accomplish?

Connell: I study school violence in general, so I’ve always been interested in the safety of students at school. As a society, we have prioritized the school building as a place where we want our children to be the safest.

I was struck after Sandy Hook, when we started seeing a lot of headlines about the number of school shootings. Around 2014, I would see headlines with really high numbers of school shootings. I thought, “Well, that can’t be true. I can’t imagine I’m missing that much.”

As a researcher, I take pride in trying to know what’s going on in my field, and so I started digging a little deeper. I started gathering all of the different school shooting incidents that were being added up to make these headlines, and what I saw is that there was a real misinformation campaign going on. When anybody hears the term “school shooting,” they think about the really tragic events, like Parkland in February, or Columbine, or Sandy Hook.

As a parent or a community member, if somebody says there’s been 300 school shootings, in our heads, we immediately think there have been 300 awful, tragic, worst-case-scenario incidents. But that’s not what was happening. It struck me that there was a lot of confusion, and I decided that nobody was doing a good job of trying to understand these incidents, either as a whole or as individual typologies.

I worked with some graduate students and some undergraduate students, and we started compiling this database. As we did, we were more and more struck by the different types of incidents that make up school shootings. In fact, as I expected, the vast majority of them are not rampage shootings. They are the same kinds of gun violence incidents that we see in the rest of our communities.

I started working with some scholars in the field who had expertise in how to collect these type of data. Getting that kind of information is really difficult, as you can imagine, because it encompasses so many things: It’s information about an incident, a victim, an offender, a school. That’s how we teamed up with Michigan State and John Jay University and the University of Maryland to come together and start doing this at a level of detail and completeness that was only possible when we got federal funding and support.

A lot of school shootings that are being categorized by, say, Everytown for Gun Safety, are not necessarily what most people think of as school shootings. In fact, the Everytown tracker includes incidents that do not result in injury. Why is it important for you to make that distinction?

One of the things that interests us is intent. Our database currently only looks at fatalities or injuries. We do not include accidental discharge unless someone is injured.

We have to be reasonable in our starting approach so that we’re not overwhelmed. But also, without injury or fatality, you are much less likely — especially as we go back historically — to get a count. From our perspective, it’s realistic that if we were to include non-injuries, we would only get the more recent ones as people have become more alert to this issue and as the media has publicized them. That is not a fair comparison. That would accidentally or unintentionally suggest an increase in events that we have no comparison point to show if they had increased. That’s a very real concern of mine, and that is why we are not looking at cases where there was no injury or fatality.

At the beginning of 2018, The 74 began tracking shooting incidents in schools. Education Week started a tracker around the same time, and their numbers look a little different than ours because we use slightly different methodologies. Our tracker includes universities, for example. What specific parameters did you put into place as you looked at these incidents?

The parameters for us are: There must have been an injury or a fatality, and we are very broad in how we describe injury, even unintentional injuries.

It must be K-12, and the reason it must be K-12 is because those are the school grades that are the purview of the government.

We absolutely include private schools. But a preschool is not overseen by a government agency. It’s not their responsibility. It’s not something everybody has to send their child to, whereas K-12 is statutorily mandated: You must send your child to school unless you have a homeschool exception. To us, everybody in the United States benefits from knowing that those places are safe.

We include incidents or events that the school has control over. A K-12 campus would not include the administrative building where the superintendent is and where there are no children, because that’s a different risk. But it would include the bus. The bus ride is under the control of the school. The school’s rules matter. After-school activities matter. We have lots of instances of extracurricular activities that lead to shootings. Oftentimes, those shootings are perpetrated by people who are not even members of the school community.

We spent a lot of time, especially for these off-campus events — and there aren’t many — wondering, “Does the school day end at 4 o’clock, and therefore does anything that happens after 4 o’clock count?” We decided it does, because we know those incidents are getting wrapped up into the conversation. We can all agree as a society that keeping the school safe even when children are not there is probably a good thing.

Your tracking will go back to 1990, which is a lot of data to analyze. Where are you at in that process, and are you able to discuss any trends that you’ve found so far?

We were just awarded this grant last year, so we have not even finalized our data collection process. We are not scheduled to be finalized until December, although we do have very high hopes to be really close by the end of the summer.

What we have found is that it’s a lot harder, as you could imagine, finding information on the incidents from the 1990s than it is from 2010 and beyond.

In terms of trends, there are a couple of things that are important to keep in mind from the perspective of policymaking. One is that, as of now, we don’t think there is an increase in the number of incidents as much as there is an increase in the attention to the incidents.

Anything before Columbine, there’s a real reporting bias that we may not ever be able to truly overcome. So I really caution people when they say that school shootings have increased, because I don’t think that’s going to be our finding. Certainly, attention has increased.

The other trend that is really important to keep communicating is that the number of rampage-like incidents remains extremely low, and they are a relatively small subsection of the shootings we are analyzing. That’s really important because those are the kinds of events that create the most fear — understandably so. But they may not be the ones that should be driving all policy.

From a policy and a prevention perspective, there are several different typologies of shooting incidents that need to be considered. Even something as simple as whether or not the perpetrator was a student or an adult has very different policy implications. The same prevention implications will not exist for each typology.

All of these problems are not going to be solved the same way. School shootings are much more complex than we typically understand, and our solutions are going to need to be much more complex than the kind of conversations we’ve been having about them.

I give the Parkland students so much credit, but one of the things they have done in standing up is to bring attention to young people in low-income, generally minority, communities who have been plagued by gun violence in their schools and in their communities for a very long time. That plays out in our data set: Communities that have problems with gun violence also have problems with guns in their schools. So that is not a school-based solution. That is a community-based solution. Your school is just another place in the community.

That is another thing that is getting lost in our discussion about the very rare, absolutely 100 percent tragic mass shootings that we want to prevent: From a policy perspective, there may be other things that we can do sooner, more easily, to prevent what, in many cases, is just the more common gun violence that shows up at a school.

If you could reshape the narrative about gun violence in schools, how would you do it?

I would make it more inclusive about general issues surrounding victimization and violence in schools and around school communities. Schools are the safest they’ve ever been. That’s really important to say as much as we can. But there’s still lots of room for improvement to ensure that everybody has a safe and healthy experience.

We have no reason right now to believe there is a relationship between being victimized in school and being a school shooter. We don’t really think that’s a thing. But we do think there’s a relationship between being victimized in school and a lot of other negative outcomes. Based on our data, we also know that many of the school shootings we see are suicides. They are not suicide by cop, but they are suicides at 2 o’clock in the morning when a young person, a student, goes to the gym and shoots themselves and is found the next morning.

That tells me that there are other things we need to be doing to protect those children, and so that conversation about victimization and violence matters. It also matters because we know that students in school who are violent, who are bullies, who are aggressive also go on to have pretty bad life circumstances. If we intervene on both sides, we’ve got so much more potential for long-term positive effects.

I don’t want our focus about school violence to be just on these rare events because, in doing so, we lose a lot of our ability to make the entire experience safer, healthier, and better for students across the board.

What indication do you have that bullying isn’t a motivator for school shootings?

If we look at the Columbine and Jonesboro perpetrators, the narrative surrounding them started that they were bullied. But if you actually look at the case studies, these were not kids who were bullied and sitting alone in the corner and didn’t have friends and decided to turn toward this event as a way to fix that or to fight against that. These kids, while they might not have been the most popular, were relatively middling in the popularity spectrum. Did they get teased by their peers? Yes. But they did just as much teasing as they received.

So these were not passive victims, and in some cases might have given more than they received. Or they were students who were acting in bizarre ways through, for example, obsessions with being a Nazi. So it’s not outside the realm of normal that that got a reaction from their peers.

That’s a lot different than bullying somebody because they’re different, because of a disability, or poverty. These were kids who were standing out on purpose for bad reasons, and then they were mad.

So in those instances, there’s a lot more going on. Bullying is a real small part of the equation. That doesn’t mean that there aren’t children who, because they’re victimized, act out in some way. That’s why that broader conversation about school violence and school victimization is so important and it’s so vital that we keep having it.

What kinds of trends are you seeing play out in the behaviors of shooters?

It really depends on what kind of shooting incident you’re talking about. When we look at these rampage shootings, there appears to be evidence of some types of serious mental illness, specifically schizoaffective disorders.

Rampage shootings are a very, very small subset. It’s also difficult to make comparisons for a variety of reasons — the biggest being that many of the perpetrators die during the shooting, so we only have interviews with those who have survived. That is a very biased sample.

Columbine and Jonesboro are two very good examples of young people who probably had some type of early-onset schizoaffective disorder that distorted their thinking to the point that, for these individuals, a school shooting seemed like a reasonable thing to do.

There’s also a high number that are suicide only, and that the only intention ever for the student is suicide. It’s not to take any victims along with them.

There’s a bunch that are adult-perpetrated, and they look just like all the other adult-perpetrated homicides. There was a case in Florida where a janitor killed his two co-workers. That’s not a school shooting. That’s workplace violence. Or the San Bernardino case last year: The man followed his estranged wife to work — she happened to work at a school — so he opened fire.

Why is it important to focus on school shootings, and what do you think could be learned from school shootings that could teach lessons about broader gun violence?

I focus on school shootings because I focus on school safety. This is a piece of school safety, and to me it was a piece of school safety that was getting turned into something that I didn’t feel was representative of what was happening.

I do not know what kind of effect our research will have on the conversation about gun violence in the United States. It never occurred to us that it would be part of this conversation. I continue to really hope it will be part of the conversation about school security because one of the things that happens after tragic events is this sort of lockdown mentality.

Imagine being a young person and all of a sudden your school has security cameras or a metal detector. Or if you’re a teacher and you show up one day and there’s a whole new set of rules, and you’re like, “How am I supposed to incorporate this?”

We’re having this conversation about school resource officers and police in schools even though there is very little evidence to suggest it’s a good idea. Most of the evidence on school resource officers says it’s probably not a good idea and, in fact, in many communities it is increasing the criminalization of youth.

The school-to-prison pipeline is real. My hope when I started, and as I continue to do this, was that we would be able to give real answers to law enforcement and communities and school administrators about what are the actual patterns they have to worry about.

The way I see it, a gun should never be in a school. There is not a state that says it’s OK for somebody under the age of 18 to buy a gun and bring it to school with them. So that debate has been had. We’ve already agreed that these should be gun-free schools. Now, recently, people have started to say things like, “Let’s put guns in the hands of teachers.” I hope that the collective crazy that led to that will go away. But it’s not up for debate that 10-year-olds should have guns. What is up for debate is, “How do we make sure that if a kid picks up a gun from mom or dad or grandpa or grandma, or steals it from the house down the street, that they can’t get into the front door of the school with that gun?” Because there’s lots of room to suggest we could probably improve on that. That’s where I see us really being able to answer questions like, “What are the ways we can think about safety and school space in meaningful ways that do not turn our schools into maximum security prisons?”

From the research that you’ve done so far, what would be the most effective method to stop the lion’s share of the problem?

Probably social-emotional learning — whole-school-centered approaches to improve climate, clarify expectations, and support teachers and administrators in creating a community of trust and support.

I say that without even looking at my database, because we’ve seen those across the board being the most effective kinds of programs to decrease victimization and violence in schools.

One of the things we hope the data does is inform the use of what we call crime prevention through environmental design. We use this a lot in our public spaces. When we look at the use of educational space, a lot of what we have focused on is learning. We believe that learning and safety don’t have to be mutually exclusive. The ability to give schools non-security-feeling tools to increase and create safety is also pretty meaningful.

One of the problems in Columbine, if you look at the schematics, is that it basically created a spot that when the fire alarms went off, the students started running out, and they all had to run through the same place. In doing so, they became very easy targets to anybody who knew even a little bit about tactical strategies. Those are the kinds of things we can hopefully get a better handle on and find better ways to improve upon in the building and refurbishing of school buildings. We need to meet educational needs but also safety needs in ways that are warm and inviting and welcoming, not ways that require metal detectors and cameras and these very prison-like experiences.

Let’s treat students like they are very valuable parts of our society, which they are. Let’s not make them feel under scrutiny because they’re doing the thing we tell them they have to do, which is go to school and learn so that they can be better people.

Get stories like these delivered straight to your inbox. Sign up for The 74 Newsletter