Profiling Bad Apples

Should schools be in the business of predicting violent behavior among students? by SCOTT LaFEE

It's a roll call most school administrators sadly know all too well.


  • Barry Loukaitis, 14, a student at Frontier Junior High in Moses Lake, Wash. On Feb. 2, 1996, Loukaitis shot and killed two students and a teacher on campus. He is now serving two life sentences.

  • Luke Woodham, 16, of Pearl High School in Pearl, Miss. On Oct. 1, 1997, Woodham stabbed his mother to death, then went to school and fatally shot two students. He's serving three life sentences.

  • Michael Carneal, 14, of Heath High in West Paducah, Ky. On Dec. 1, 1997, Carneal killed three students at school. He's serving a single life sentence.

  • Andrew Golden, 11, and Mitchell Johnson, 13, both students at Westside Middle School in Jonesboro, Ark. On March 24, 1998, they fatally shot a teacher and four students. Both boys are confined to a juvenile detention facility but will be released at age 21.

  • Kipland Kinkel, 15, attended Thurston High in Springfield, Ore. On May 20, 1998, Kinkel shot his parents to death. The following day he went to school and murdered two students. He was sentenced in November to 112 years in prison.

  • Eric Harris, 18, and Dylan Klebold, 17, both from Littleton, Colo. On April 20, 1999, they launched the bloodiest school assault of all, killing one teacher and 12 students at Columbine High School. Both committed suicide immediately after the murders.

Six school shootings with multiple fatalities in less than two years. Each event was different, yet each provoked the same powerful emotions--horror, grief, despair--and a singular question: Why? What turned these students--some barely into their teens--into killers? Were they alike in some way? Did they share characteristics that should have identified them to school officials and authorities as would-be murderers?

A Controversial Tact
In the aftermath of Moses Lake, Jonesboro and Littleton, with fears of more tragic acts and the public clamoring for action, educators across the country have struggled to boost security and restore the notion that school is a safe place to be. Some efforts have been obvious and material: more uniformed guards on campus, new metal detectors at school entrances, mandatory student identification badges, stricter dress codes, panic buttons in classrooms and surveillance cameras in the hallways.

But a growing number of school administrators are attempting to grapple with that harder task of developing programs and procedures that will identify violence-prone students before any more bullets fly on school grounds.

The idea is called student profiling, an admittedly charged term with negative connotations arising from past abuse involving law enforcement agencies and racial bias. The FBI, a recognized authority on the subject of profiling and its use, prefers an alternative phrase: behavioral assessment.

Whatever the label, though, the idea of psychologically evaluating and treating students who may become violent is clearly controversial.

"Is profiling a good or a bad idea? I don’t know," says Ronald D. Stephens, executive director of the National School Safety Center in Westlake Village, Calif. "What I do know is that a lot of administrators are talking about it. And that if it is done, it must be done with a great deal of caution."

Bright Signals
The fact school leaders are considering such unconventional measures as profiling comes at a time when the state of school safety is largely positive.

A recent study in the Journal of the American Medical Association noted a 30 percent drop from 1991 to 1997 in the number of students carrying weapons to school and a 14 percent drop in student fights. The number of killings on school campuses also has fallen, according to the National School Safety Center, from 54 deaths in 1992-93 to 24 in 1998-99.

Contrast those numbers with these: Almost three kids die every day on average from family abuse and nearly nine per day in car accidents. In fact, killings on school property account for less than 1 percent of all homicides of children.

Yet life on most school campuses is decidedly more complicated and dangerous than it was just a couple of generations ago. In 1940, for example, public school teachers rated among their top disciplinary problems talking out of turn, chewing gum, making noise, running in the hall and littering, according to the Reason Public Policy Institute. In 1990, that list featured drug and alcohol abuse, pregnancy, suicide, rape and assault.

While the rate of juvenile violence overall is lessening, the number of school-based incidents in which more than one person is killed appears to be growing. Between 1992 and 1995, there were just two shootings with multiple fatalities. Since late 1995, 12 have occurred. The only comfort--and it’s not much of one--is that most observers of school safety say it is premature to conclude whether the increase marks a trend or is simply a statistical aberration.

A Preventative Tool
Of course, any academic debate about trends and aberrations becomes irrelevant in the wake of iconic events like Paducah or Littleton. When innocent children and teachers die on campus, frightened parents, teachers and others demand immediate action from school officials, police and politicians. This is when proposed measures such as student profiling come to the fore.

"The thing that drives ideas like profiling is the fact that historically the villain has always come from without," says Peter Blauvelt, a school security consultant and president of the National Alliance for Safe Schools in College Park, Md. "What’s happened with these shootings is that these kids were already inside the school building. They were us. And it’s a terrifying idea to think you or your child may be sitting next to a potential killer."

Blauvelt broadly supports the idea of profiling, though he prefers to call it "incident" profiling. "It’s my contention that if you sufficiently monitor, say, the number of student fights, where they happen and on what days, you’ll end up with a pretty good idea of where to anticipate that problem.

"Doing this is a preventative management tool," adds Blauvelt, former director of security in the 125,000-student Prince George’s County, Md., system. "By expanding the profiling of incidents to include student referral data, you can find out why kids are sent to the office. Are there commonalities among the offenders? Do you, as an administrator, tend to wind up with certain names and particulars?"

Most school districts right now don’t really review this incident material, even though most collect it to some degree. Reasonable educators just need to develop better internal abilities to monitor and assess what’s going on, he says.

"Sure, there’s potential for abuse here, but in education, there’s that potential for a lot of things," Blauvelt notes. "People just have to be careful not to cross the wrong line."

A Cautious FBI
In recent years, various organizations, from the U.S. Department of Education to an ever-growing array of private consulting firms, have drafted and issued guidelines to help school officials identify troubled students.

Generally, these lists of behavioral warning signs are similar, identifying students as potentially problematic if they are, among other things: chronically depressed, uncontrollably angry, living in a broken or abusive home, have a history of violent behavior or threat-making or have easy access to weapons. All of the school shooters in the last two years exhibited one or more of these traits.

Thus, the thinking goes, these lists offer insight and guidance to educators looking to spot trouble before it happens. However, even the most ardent of list advocates caution that they are just a place to start. The key to successful intervention before a troubled student turns violent, experts say, is what happens after that student has been identified.

"Basically, I tell school officials and others attending seminars on violence that there are warning signs, but you need to do more than just look for them," says Terry Royster, a supervisory special agent who works in the FBI Academy's Behavioral Sciences Unit. "You need to hook up with resources that will help with what happens next."

A thorough, accurate behavioral assessment is complicated, time-consuming and requires experience at criminal investigative analysis, says Royster's colleague, FBI Special Agent Joseph Harpold. Few school administrators can claim that sort of expertise on their staffs.

But Royster and Harpold insist educators needn't feel helpless. Rather, say the FBI agents, educators should play to their strengths and seek competent assistance where they need it. "What I stress is to really forget the school shooter behavioral assessments and go into the classroom," Royster says. "Every teacher can tell you who's likely to cause trouble. It's common sense. Young children in kindergarten or 1st grade who have histories of tantrums or uncontrollable outbursts of anger are probably the same students who will pose the greatest threat to themselves and others later if their problems aren't addressed early on."

Applying Diagnostics
Steve Balen, superintendent of the Granite City, Ill., Community Unit School District, across the river from St. Louis, knew he needed to do more to protect the district's 8,400 students and hundreds of teachers and staff after the 1998 shootings at Jonesboro.

"We got a list of warning signs, which was nice to have. But frankly it meant nothing if we didn't also create a vehicle for acting upon these signs," Balen says. "It made sense to apply a diagnostic system to the problem, to establish a group of professionals to receive referrals from staff or parents or whoever. As a school district, we have the means to require people to take this issue seriously."

Balen, who believes his district was the first nationally in February 1998 to establish a written policy for psychologically assessing potentially violent students, crafted a two-step procedure.

The initial step requires teachers and staff to immediately report to district officials any behavior or action by a student that might lead to armed violence. This information, along with copies of grades, past behavior reports, medical records and other material obtained from outside sources, then is forwarded to an evaluation team whose members include a school resource officer or counselor, a social worker, a psychologist and an administrator.

"The team always has at least three members who are not part of the school district, who can look at the issues from a distance. Everything is thrown on the table for evaluation," says Balen. Neither the student nor his or her parents are informed or involved in the process at this stage.

"Now maybe the problem is resolved at this level," the superintendent says. "Maybe the team says there really is no problem. That's great. But if there is a problem, we go to level 2. That's when the parents are approached. We ask for permission to do a psychological evaluation. We go directly to the student. A social worker visits the student's home to see if perhaps we have a situation like that at Columbine where the parents seemed to have been oblivious to their kids building bombs in the garage. The social worker also looks for issues like access to guns at home or within the family."

Findings from the second stage are passed to the school board, which can order the parents to pursue remedial treatment for the student. If they refuse or fail to comply, the superintendent says the student would be placed in an alternative educational program of the board's choosing. Expulsion is used only as a last resort.

"All we are looking to do is get a troubled student to the right kind of help," says Balen, noting the district has used this process only three times, with none reaching level 2. "We want teachers, staff and others to relay their suspicions and observations. If they do that, we'll take it seriously and make the appropriate referrals."

Questionable Merit
Not everybody thinks student profiling is an unabashedly good idea. Some wonder if it even works. After all, descriptions of moody, angry, confrontational and low self-esteem can be used to describe almost any teen-ager at some point in their lives.

"I don’t think we have any data to show whether it is effective or not," says Lois Flaherty, a practicing child and adolescent psychologist and spokesperson for the American Psychological Association. "And the lack of research is just one of many issues here. There’s the question of who is doing the identifying of students and the evaluation. What happens with the results? Will they be used to single kids out for further stigmatization and isolation? What are the civil liberties concerns?"

In fact, the American Civil Liberties Union already has voiced its skepticism to student profiling, noting that some educators, though well-meaning, might misuse or abuse the idea. "In some cases there's not a whole lot of thought going into it," said ACLU spokesperson Emily Whitfield.

George Woons, superintendent of the Kent Intermediate School District in Grand Rapids, Mich., shares at least some of the ACLU's concerns.

"I think [student profiling] raises all sorts of constitutional issues," says Woons. "I wouldn’t advocate it at all. I don’t think you have the legal right unless you get a parental sign-off. When you get into something like this, boy oh boy, it’s a lawyer’s dream, a real invasion of privacy.

"And once you have [a profiling program in place], what do you do with it?" he asks. "Once you have it, if you don’t do anything with it, and a kid does something terrible, well, it’s a whole new can of worms. Let’s say a profile showed a kid to be aggressive. A counselor says it’s not serious, but then the kid blows up the restroom. You have this assessment instrument that provided cautionary information. Somebody’s going to ask, ‘Why didn’t you act on the information?’"

Still, Woons recognizes that educators cannot simply sit back and wait for the next school shooting tragedy. He has made eight recommendations in his district to improve school safety. In some ways, they are not much different than many profiling programs promoted elsewhere. Among the recommendations:

  • Develop procedures and training for reacting to critical incidents;

  • Establish school resource teams composed of staff, mental health officials and law enforcement agencies to review ways to respond to reports of potential problems;

  • Make school resource officers more visible, such as employing more uniformed security guards;

  • Improve communication with students by, for example, creating a 24-hour, anonymous hotline; and

  • Form better one-on-one relationships between staff and students.

    "It’s a sort of holistic approach," Woons says. "But for any of this to work, it requires the help and participation of everyone: students, parents, teachers, administrators, clergy, police, the whole community."

  • A Social Imperative
    Experts caution that lists of warning signs or subsequent psychological evaluations are not foolproof. "The science here is not such that we can identify very accurately kids who will become violent," says Michael Greene, executive director of The Violence Institute at the New Jersey University of Medicine and Dentistry.

    But that imprecision does not dissuade administrators like Mary Leiker.

    "I have no qualms with doing something like profiling," says Leiker, superintendent of the 8,600-student Kentwood, Mich., Public Schools. "It's what we do every day as educators. We assess children. This is not about doing the actual psychoanalysis. This is about looking for evidence that suggests more needs to be done, that a particular student needs to be evaluated by professionals. It’s simply another service that public schools are going to be expected to provide."

    Leiker has installed an assessment program in her district similar to Balen's in Granite City. It too consists of an initial review to determine whether a particular student poses a credible threat, followed by a professional evaluation and referral.

    "We want the student to get serious intervention," Leiker says. "If it’s something beyond the scope of what we can do, we’ll try to help the student and parents get help elsewhere, even if we have to pay for it. Only if the parents do not act or refuse to help do we suspend or expel the student and contact the police and social services."

    In the end, she says school administrators everywhere may wind up doing some sort of student profiling for violence, no matter what their qualms. It will be a social imperative.

    "Profiling isn’t something most of us think we're going to do," adds Leiker, who has been in school administration for 25 years. "But when you think about the top issues of education, creating safe schools is one of the biggest. Profiling doesn’t guarantee absolute safety. There is no 100 percent accurate predictor, but it can be an effective tool.

    "And the fact is, I have to live with myself. If I, as a superintendent and educator, left one stone unturned in trying to keep children safe, if I lost one child because of it, I don’t know how I would cope."

    Scott LaFee is a science and health writer at the San Diego Union-Tribune. E-mail: scott.lafee@uniontrib.com