The week before winter break, snow is piled up around St. Louis Park High School, a low-slung, rambling brick complex in suburban Minneapolis. And more snow is falling.

This is a big, diverse school with proud roots. Alumni include Joel and Ethan Coen, who shot their semiautobiographical 2009 drama, A Serious Man, in this area, once a Jewish enclave, which today has immigrants from all over the world.

But in 1998, when Angela Jerabek was a school counselor for freshmen here, she was "discouraged." For five years running, about half the ninth-grade students had been failing at least one course.

"I went to the principal to say, 'I don't think I'm doing this job well and I feel like I should resign.' " Her principal had a different idea.

"He said [the problem] wasn't just with me or just with our high school. That this really was an issue that was occurring in high schools across the country, and that we really needed to look at a new solution."

And he challenged her to come up with one.

One of the biggest questions in education is whether it's even possible to turn around a low-performing school. The Obama administration spent $7 billion on school turnarounds, and their final verdict was that nothing really worked.

The solution Jerabek developed is called the BARR method, for Building Assets, Reducing Risks. And, she has an unusual amount of evidence. Out of 172 projects in a big federal innovation grant program, BARR is the only one that progressed through randomized controlled trials to win grants at all three levels: innovation, development and scale-up.

The ninth-grade shock

BARR doesn't require changing the teachers or the students in a school. It doesn't overhaul the curriculum or discipline. It doesn't require flashy technology. It's based on something simple, and decidedly unsexy: meetings.

Jerabek decided to focus her efforts on freshman year. Many students stumble in the transition to high school — more than any other year of school. A failing grade in that first year significantly raises your risk of dropping out. And that, in turn, can shadow your entire life. Just from one F grade. Researchers call this the "ninth-grade shock."

So, Jerabek figured out a way that all the adults in a school building could come together to try to cushion that shock.

Up on the third floor here at St. Louis Park, there's a room with no windows, but inspirational posters and free candy on the table. We're here for the weekly block meeting. Here, teachers who share the same group of students sit down over a shared Google doc with the school support staff: the social worker, the counselor, sometimes even the police officer assigned to the school.

The core of the BARR method is these meetings, where educators talk about three things. First, the data — on attendance, behavior and grades. Alex Polk, a science teacher, observes of one student: "She's been late at least 19 times and is late with the same group of people each time."

Second, the big picture of the students' lives. Daniel Perez, the school social worker, shares his notes from meetings with students: "He and his mom don't get along, he feels ignored by her, Dad is not in the picture, he does not have friends here at Park and he is not interested in making friends."

Finally, they come up with a personalized plan for what to do to help each student, like signing them up for tutoring, calling home or holding a mediation.

Tjessa Arradando, a sophomore who loves chemistry and writes for the school paper, explains how BARR worked for her.

"I have anxiety, so if I was stressed I'd go down to the counselor's office and they would kind of talk to me, come up with a plan on how to, you know, adjust at certain times." She had a pass to get out of class; the counselors let the teachers know about her needs, so Arradando felt like all the adults were on her side. This year, she says, she's managing much better.

Building a fuller picture of a student

In most schools, teachers meet regularly to talk about subject areas or administrative duties. In BARR schools, the point of block meetings is to pool knowledge on students. Most teachers only see a student for no more than 45 minutes a day, Jerabek points out. They may not have the background knowledge to understand what is really going on when a student is acting out.

Sometimes, piecing together information creates a startling picture.

At a BARR school in California, one teacher raised a concern about a specific group of 14-year-old girls that tended to sit with older boys rather than other freshmen at lunch. A second teacher noticed that the same girls were missing class on Mondays and Fridays. A third teacher was concerned about dress code violations. No one was particularly alarmed, but when they cross-referenced the information and followed up, Jerabek says, they discovered that the girls were being trafficked for sex.

There's a second BARR meeting each week called Risk Review, about the kids in the toughest situations. Just on one Thursday, we hear about drug issues, fears of deportation, chronic illness, eating disorders and a student who may not have a place to sleep.

Even though BARR, by definition, focuses on students with problems, it also highlights strengths. Kelly Brown, the BARR coordinator at St. Louis Park, probes teachers again and again to name what's going right: like a student with a good sense of humor, say, or a supportive parent.

This is deliberate, says Jerabek: "To be able to identify the strengths does require that you've actually built a relationship with a student."

There's a classroom component to the BARR method as well, a social and emotional curriculum called iTime. It allows students and teachers to get to know each other better and, again, build positive relationships.

For example, in Kara Cisco's civics class on the day we visit, students are giving each other awards based on the Preamble to the Constitution. "Promote the general welfare" might be a student who shares their snacks.

This combination of hard data and soft skills has paid off. In the first year of BARR here at St. Louis Park, the failure rate for freshmen was cut in half, from 50 percent to 25 percent, and those results stuck.

Gold-standard evidence

As BARR spread, researchers tested the model by the gold standard: randomized controlled trials. And they found over and over again, it raises test scores and pass rates, reduces absences and suspensions, and students feel more engaged and challenged at school. The effects are particularly large for students of color, male students and students from low-income families.

Across big-city schools like St. Louis Park, there's a 40 percent drop in failure rates from BARR, and in small rural schools it's 29 percent on average.

The principal here, Scott Meyers says the BARR model does cost some money, because of the adjustment in teacher schedules. And it may be harder to put in place in districts that are really hurting for support staff like counselors. But, Jerabek likes to point out, BARR shows you can make big changes without changing the students or the teachers in a school, just by focusing on relationships.

"Many students, they just work harder when they know that the adults care about them."

Copyright 2018 NPR. To see more, visit http://www.npr.org/.