When “Tara” dropped out of high school to have her baby, few of us were truly shocked. All of her teachers had seen the warning signs along the way: her troubled home environment, her slipping grades, her physical changes, and, of course, the increasing teasing she endured. We could see with 20/20 hindsight the road that led to that day.
It was so easy in retrospect to diagnose what went wrong, but surely there could have been a way to avoid this, we reasoned. While Tara was far down her path toward dropping out when she became my geometry student, the symptoms must have been there in middle school or even earlier.
As I looked back at the students we lost—to juvenile detention, pregnancy, or simple academic failure—I couldn’t help but wonder if we were missing something. There had to be some kind of tool to help figure out whether students were on track for dropping out before they were at imminent risk.
It turned out there was a model for that kind of risk prediction. But it was in the most unlikely of places.
Right now, we’re relying on teachers and parents to observe, analyze, and make decisions, but we're not giving them the tools to go beyond anecdotes and bias."
Five years ago, authorities in Maryland were facing a very different issue, with homicides related to domestic violence. Women who had been repeat victims of domestic violence were all too frequently being murdered by their partners, and police were at a loss for how to stop it. Police officers simply couldn’t tell which cases were more likely to turn deadly. They had an equally hard time understanding why so few of the women who called them—a paltry one in 25—decided to go to a shelter, despite the fact that it improved their survival odds by 60 percent.
In 2007, the state was graced with a solution that seemed almost too simple. Jacquelyn Campbell, an advocate for reducing domestic violence, and a team of researchers had combed through mountains of domestic-violence data and found a small number of key indicators that dramatically increased the likelihood of domestic homicides.
The team developed a simple list of 11 questions for police officers to ask at the scene of any domestic dispute, a “domestic violence indicator.” Answering yes to any one of the first three indicator questions (the most important predictors), or any four of the remaining eight questions, automatically placed a woman in a high-risk group for potential homicide. Police officers were then instructed to give the woman a verbal assessment explaining her risk and connect her to the nearest shelter by handing her a police phone (so the number could not be tracked). If she chose not to accept the help or stay at a shelter at that point, that was up to her.
What kind of impact has this simple indicator assessment had? Despite near-flat rates of domestic-violence homicides nationwide, Maryland has reduced the incidence of such homicides by nearly 40 percent over the last five years.
There’s a lesson in the success of Campbell’s team that could have helped Tara—and could still help millions of students like her.
Students often exhibit signs very early on that they are disengaging from learning, but the biggest barrier to spotting those signs is that teachers, parents, and others can’t tell what is part of the normal growth process and what’s a symptom of something more serious. Much like with the domestic-abuse victims in Maryland, however, a select number of leading indicators can predict whether a student is likely to have a problem in school further down the line. And there’s likely a similar solution to be found in a simple assessment for students.
We need clear steps that parents and teachers can take when they identify high-risk indicators in a student."
By developing a set of questions based on key indicators (many of which have already been widely researched and published), we can equip teachers, parents, and others to assess the likelihood that a student is going to become disengaged and to help solve problems before they become intractable. We would know whether a particular kind of behavior is likely an act of harmless youth rebellion or part of a serious trend toward more dire and harmful behaviors down the line.
And similar to the Maryland model, asking the questions and assessing the risks aren’t enough. We need clear steps that parents and teachers can take when they identify high-risk indicators in a student. Many of those actions likely exist already—efforts such as tutoring programs, after-school activities, support groups, and even alternative education paths—but the challenge lies in linking the right students to the right opportunities when it will make the biggest difference.
Right now, we’re relying on teachers and parents to observe, analyze, and make decisions, but we’re not giving them the tools to go beyond anecdotes and bias. Relying primarily on the professional judgment of adults in the school system is unfair, both to the adults and the students, particularly when they often have access only to fragmented and partial information about students.
If by implementing systemwide indicator-assessments we could see even half the impact on education engagement as Maryland has seen in its domestic-homicide rates, we would effectively prevent a quarter of a million students annually from dropping out of school. And we would help keep scores more from disconnecting with education in less obvious and detrimental ways.
Disengaged students don’t need to fall through the cracks. But changing their fate requires us to approach dropping out like the systemic and measurable problem it is—and facing it with a systemic and measurable solution. I know it would have made a world of difference to Tara.
