Close to Home


January 20, 2015


The idea for the early warning system (EWS) work came out of a meeting I had with superintendents and other leaders from the AA districts of Montana—the seven largest districts in the state. What I heard over and over was, “We have students in our districts who are struggling, and we’re providing those students with interventions. We think they’re working, but we really have no concrete evidence.“

As I listened to those district team members, my inner economist kept leaping ahead, considering the various statistical models that could address their needs—randomized controlled trial studies, a regression discontinuity design, the applicability of a fixed effect or Cox Hazard model. (Yes, it’s a sickness.) My brain raced through the options and crossed them off just as quickly: Takes too long, costs too much, isn’t sustainable. The districts would need to have years of consistent longitudinal data before they could know with any certainty that something was working, and even then it would be qualified as working “within a standard deviation of X.” It really isn’t feasible for a district to rigorously evaluate every program and initiative, and those students don’t have years to wait around for us to help them. This is urgent; these students are dropping out now. And that's when everything hit home for me.

As the only one of the five children in my family to graduate from high school, dropout prevention has always been more to me than an abstract issue or a cold set of data. The dropout problem is as real to me as the faces of my family members and the streets of my small town. As I listened to the Montana district leaders, I thought back to my own high school years. What made the difference for me? Why did I make it through, against all odds? The answer was a human one: It was the genuine understanding, compassion, and interest of a few key teachers at critical moments in my life. Without them, I most likely would have dropped out, defeated by the pressures of becoming a single teenage mother.

And yet, one thing I had learned from my educational training is that the presence of caring, committed teachers is not enough. The human touch and compassion are critical, but of equal importance are reliable data that can help them see if what they are doing is making a difference, and maybe even allow them to intervene earlier with struggling students.

I didn’t go from being a straight-A student to a pregnant teenager overnight. The beginning of my senior year, I had emancipated myself and was living with a roommate. I was staying up late, making poor choices, and began missing a lot of school. And that’s exactly where an “early warning system” could have come into play. Such a system has the potential to more proactively intervene before students fall so far off track that it takes not only a village but a sizable chunk of the village’s resources to save them.

From what I had read and heard from education colleagues across the country, early warning systems were emerging as a solid framework for turning data that schools were already collecting, such as attendance, behavior, and course performance, into real-time information schools could use to help students proactively. Thinking back to my situation, I would have been flagged pretty early on in the semester based on attendance alone, in addition to my plummeting grades.

In working closely with school staff members, I’ve seen that the challenge isn’t identifying the students who are struggling the most; they can pretty easily identify those students by that point. But, what if they had been able to identify those students earlier, right when they first started to stumble? From an economics perspective, of course, this makes sense. It’s more efficient. But, beyond that, it addresses the human desire to make a difference, as well as the equally human desire to know that someone cares.

How empowering could it be for a student to know that a large group of caring, competent adults is watching out for her? To know that, “Yes, they notice that I’m struggling. I’m not alone in my situation. They care.” And how empowering could it be to be a part of such a staff and to know that you are not out there alone, paddling for all you’re worth, trying to save every drowning soul that comes through the door, and that, in fact, your school—your entire district—has a well designed system in place for identifying these students at the earliest possible moment and getting them the exact intervention they need.

Data can give us the power to move beyond anecdotal hunches and programs that might make us feel good but that have no real impact. Data give us the power to identify more accurately not only individual students who are struggling but also the areas where the system is failing students as a whole, and even where it is failing particular groups of students.

Is your curriculum and instruction working for all students, most students, only a few students? Does it address the unique needs of your student population? Data can help us unravel this, but only if they are collected and presented in a way that is easy to translate into action. You shouldn’t need to have a degree in econometrics to be able to collect, analyze, and apply student data in a powerful and potentially life-changing way. A well designed early warning system is one way to meet that need.

Throughout the process of writing this guide, my coauthor and I continually questioned how practitioners could use the information in concrete ways. To that end, we have provided examples of actual schools and districts that have implemented an EWS with positive results. While we still lack a “gold standard” research base for EWS effectiveness, preliminary trends are significant. For example, according to an NPR report in April 2014, “a third of students from [a Florida high school district] who were flagged for missing school got back on track to graduation. Two-thirds of the students who were having behavioral problems made a turnaround.” Those are staggering numbers that point to the potential of a well planned and executed EWS.

I hope that practitioners will find the guide to be both useful and inspiring. As I sat in that meeting with Montana AA district leaders two years ago, my goal was to collaborate with them on a project that could truly make a difference in the state. What we landed on was something that I think goes to the heart of our profession, whether as practitioners or researchers: the desire to find concrete, practical, and effective ways to meet the needs not only of a select few or even of the majority, but of all students, and especially those whose needs we have traditionally addressed either too late or not all. I'm excited to be doing this work and to be an advocate for a type of system that I believe has the potential to change lives.

Read our news article about the new REL Northwest report coauthored by Sarah Frazelle: A Practitioner's Guide to Implementing Early Warning Systems. The article includes a link to the report.