Last month, New York City's Department of Education, under Chancellor Carmen Fariña, called for an end to principal-led school suspensions without prior approval—a practice that grew in popularity during the Bloomberg years as part of a focus on "broken windows," or small crimes that herald disorder. And the Los Angeles Unified School District made a similar move two years ago, when it banned suspensions for "willful defiance," punishment that had a disproportionate impact on students of color. These large cities are at the vanguard of a shift away from zero-tolerance school discipline toward less punitive strategies that emphasize talking it out and resolving disputes among students to keep them in school.

To some extent, these massive districts are rejuvenating the "whole-child" approach integral to what's known as "progressive educaticon"—a model that was once viewed as incompatible with urban school systems. The contours of this model, which is often vaguely defined as schooling that is "child-centered" and focused on "active learning," are outlined by the educator Tom Little and writer Katherine Ellison in Loving Learning: How Progressive Education Can Save America’s Schools. Little (who died last year) toured 45 so-called progressive schools in 2013 and found several consistent features: attention to relationships; the students’ freedom, within limits, to follow their interests; and hands-on, creative projects.

But despite the allure of progressive education, Little's findings illustrate the challenge of scaling the model up to districts with large, high-poverty schools. After all, the campuses he identified are small and either private or dependent on a charismatic, passionate leader. And for the most part, little convincing evidence has existed to demonstrate that progressive education is more than just an ideal—until now.

Research has long showed that stringent discipline policies are alive and well at public schools across America. As a recent report from UCLA's Center for Civil Rights Remedies makes clear, many of the country's schools are a long way off from enjoying the values typical of progressive education. That's particularly true of the second quality identified by Little—student freedom—given that the nation's schools since 2009 have, on average, reported an annual suspension rate of 10 percent, the highest it's ever been. The rate, which started steadily increasing in 1972, is based on U.S. Department of Education’s civil-rights data and pertains to the percentage of individual students who were suspended once or more in any given year; it doesn't use the total number of suspensions at a school, which could include the same student numerous times. African Americans and students with disabilities, the report shows, are suspended at much higher rates than their white and general-education peers.

Zero-tolerance policies mean that suspension is used as a consequence for infractions ranging from severe (such as weapon possession) to minor (defiance or chronic tardiness). In some charter-school networks, including Success Academy in New York and Uncommon in Newark, as well as some public-school districts—such as Pontiac, Michigan, and Saint Louis, Missouri—nearly a third of students are suspended annually, according to the UCLA study. Meanwhile, Florida as a whole has a 19 percent suspension rate. And in Texas, nearly 60 percent of students have been suspended by the time they graduate high school, according to a 2011 report by the Council of State Governments’ Justice Center. As that report documented, suspensions and expulsions can predict a cascade of poor outcomes for kids, including failing a grade, dropping out, or becoming incarcerated. (The UCLA report excluded New York City because of inconsistently reported data.)

Nationally, African American students are suspended at three times the rate of their white counterparts, creating a "discipline gap," as Daniel Losen, the director of the Center for Civil Rights Remedies, puts it. The term, he said, shows the link between discipline trends and the socioeconomic chasm in academic achievement. The discipline gap is so well-documented that the U.S. education and justice departments issued a joint "Dear Colleague" letter last January telling school systems to fix discriminatory punitive practices.

Because numbers vary widely by school district, the prevalence of suspensions appears to correlate more with policy than it does student behavior. Suspensions have come to serve as a proxy for school climate—and on campus, climate, as anyone who went to school knows, can be warm and embracing or severe and righteous, depending on the district, school, and even classroom. "No-excuses" charter networks, such as Uncommon, Achievement First and KIPP, have managed to get low-income students to graduate and enroll in college at higher rates than the national average. Yet as Sarah Carr reported in The Atlantic last year, the strict discipline policies of these networks, whose methods have filtered into many urban-district schools, alienates many students. And some experts have concluded that these approaches are unnecessary.

In the stereotypical urban school, fights break out in hallways and hair gets set on fire in the stairwells (I remember both happening at my urban public school in the early ‘90s). Leaders at these schools, in turn, may feel they have no choice but to institute tough-love rules.

The Princeton researcher Joanne Golann observed an unnamed no-excuses urban charter school in the Northeast over the course of a year and a half between 2011 and 2013 and is slated to share her results in a paper that will be published next month in Sociology of Education. She cites one middle-school principal who witnessed the before-and-after of a strict discipline policy, explaining, "we had students who tried to burn down the school, students who brought weapons." When the principal installed typical no-excuses rules—mandates that students walk in straight lines between rooms or sit in silence if a teacher raises two fingers, for example—the atmosphere of the school apparently calmed and test scores went up. The principal concluded that "what works in urban education are rigid structures and hierarchical relations," Golann said.

But Golann points out that no one actually knows for sure if zero-tolerance practices are critical to the schools’ success. "The schools believe that they are," she said. "But actually the research evidence is much more mixed." Any positive impacts of no-excuses schools may lie more in supplementary features such as longer school days and intensive tutoring.

Getting suspended for minor offenses may even be counterproductive if a school's goal is preparing kids for college and their careers: The practice enforces obedience more than the kind of independent thinking valued by four-year colleges, according to Golann. KIPP has even publicly expressed concern that it might be able to get low-income kids to college but not keep them there at rates comparable to those of higher-income students, in part because they need to develop the self-advocacy skills familiar to children who have been taught to negotiate for themselves throughout their lives.

And administrators’ beliefs about how to control students often determine the number of kids who get suspended rather than the actual misdemeanor count, according to some studies. In 2008, principals in Indiana took a survey about their thoughts on discipline. Russell Skiba, the lead researcher, found that principals who favored zero-tolerance approaches over preventative ones suspended and expelled students at higher rates. While this may not be surprising, what was striking was that this belief in zero tolerance outweighed other factors, such as poverty or the type of infraction; the only other variable that strongly associated with suspension rates was race. "Administrators don't suspend kids because they love kicking kids out of school," Skiba said. "It happens because they don't know what else to do." In other words, if you don’t send a message that the student caught smoking in the bathroom needs to cool off at home, what other options do you have?

It turns out that there are plenty of options, and that's where progressive education steps in. Indeed, many of these options hark back to the era of early 20th-century schooling, when educators pushed back against the teaching philosophies typical of the 1800s: an emphasis on rote memorization and physical obedience, for one. "Talking only when called upon to recite, teaching where the teacher did the thinking; these conditions have meant and will always mean an imposed discipline, an imposed routine, whereas real discipline is a personal thing," explained the New York City principal Angelo Petri in his 1917 memoir A Schoolmaster of the Great City.

Self-proclaimed progressive schools deemphasized testing and discipline, replacing those practices with student-driven, hands-on learning; collaboration among schools and families; and social-emotional well-being. Petri for his part focused on making his Lower East Side school a community center for families. And that philosophy has stuck in some circles, with experts such as Little arguing nearly a century later that while progressive education’s hallmark strategies may sound "touchy-feely," they in fact lead to huge student gains and require rigorous planning.

The problem is, until now evidence demonstrating their benefits has been largely anecdotal. It's hard to pin down the gains that come with these settings: an eighth-grade class at Cambridge Friends School that's producing a magazine with younger students’ work, for example; or the third-graders at Oakland’s Park Day School who are writing notes in a confidential book to express worries about conflict in class; or Boston’s public Mission Hill, where a teacher ties green string between children's pictures of their dreams to help them to visualize connections. Charming as they are, the absence of hard data showing these strategies are more than just a nice idea has hampered efforts to formalize these less punitive practices as district policy.

But now that large school districts are adopting similar practices, however, clearer evidence is emerging. "Students learn best when they are being actively engaged in a supportive environment, not when they are worried about getting suspended for any minor incident," says Jason Fink, a spokesman for the New York City Department of Education, about the new discipline approach. New York City, which with 1.1 million students is the largest school district in the country, may have looked for inspiration from No. 2 on the list: Los Angeles, which saw suspension rates for its roughly 700,000 students plummet by 53 percent in the two years since it banned suspensions for subjective offenses such as "willful defiance." Graduation rates in Los Angeles, meanwhile, rose by 12 percent between the 2012-13 and 2013-14 school years.

New York City's DOE is slated to allocate $1.2 million toward expanding "restorative practices," a term used to describe talk-it-out behavior interventions. In these interventions, students involved in disputes or infractions participate in developing their resolutions, which include peer mediation, restorative circles, and group conferences.

The restorative approach is already well-established in some parts of the country—and outcomes suggest that it's working. After Denver Public Schools, for example, implemented a district-wide "restorative justice program" in the early 2000s, suspension rates were cut in half over seven years and the discipline gap between African American and white students shrunk by a third, according to a study conducted by the Occidental professor Thalia Gonzalez and published in the new book Closing the School Discipline Gap.

Denver's initiative was comprehensive; personnel from district staff to teachers trained in how to use a variety of restorative techniques depending on the situation, such as one-on-one talks and group conferences. Gonzalez describes one incident in which several freshmen football players decided to throw each other in dumpsters and then grabbed unwilling students to toss in as well. Instead of suspending the students, the school held a conference involving all parties, with the offenders owning up to the harm they had done and school personnel admitting that their supervision had been lax. The players volunteered in the end to miss the homecoming game and to apologize to the entire school. The goal was to get students to reflect on the effect their actions had on others and correct them, rather to simply ostracize the miscreants. After a decade of using restorative justice, standardized test scores in Denver's schools have gone up, as have graduation rates.

Meanwhile, some small public schools in New York with high percentages of low-income students, including Urban Assembly School for Law and Justice and Humanities Preparatory Academy, have also sent almost all graduates to college (88 percent and 83 percent, respectively) with near-zero suspension rates.

As Skiba’s research demonstrates, a school leader who fundamentally understands why kids misbehave and prioritizes community building can have profound impact on how students do academically. The Metropolitan Expeditionary Learning School, a new high-performing public secondary institution in Queens where 60 percent of the 800 students are low-income, attests to that finding.

"The idea of the whole child and supporting the whole child is basic to our approach," said Patrick Finley, one of its leaders. That approach relieves the campus of "a lot of the discipline work that goes on in other schools." Students start the day in advisory groups, where they discuss common challenges, and all new sixth-graders go on a five-day bonding trip at the beginning of the year. "When kids are struggling, it’s not that they don’t want to learn; it’s that they are missing some set of skills that are preventing them from learning," Finley said. "Removing them from the classroom is not building those skills."