This article is adapted from David L. Kirp’s forthcoming book, Kids First: Five Big Ideas for Transforming Children’s Lives and America’s Future (Public Affairs).   As seasoned kindergarten teachers in downtrodden neighborhoods will tell you, in every class there are some 5-year-olds who can’t sit still and won’t follow directions, who fight with their classmates and even sometimes threaten suicide. And the teachers worry, with good reason, that these children are at greater risk of becoming pregnant or getting in trouble with the law when they’re teenagers. These kids may never have had the kind of secure relationship with an adult that psychologists regard as essential for normal development, but there’s not much kindergarten teachers can do to overcome that, with dozens of students on their hands.

The arc of a child’s life isn’t entirely predictable, of course, and the odds that children will succeed improve markedly if they can count on stable adult support. That’s what a mentoring organization called Friends of the Children has set out to achieve. The nonprofit’s strategy is simple to state and devilishly hard to pull off: start very early, in kindergarten; be steadfast; take all the time these kids need to connect with a caring adult; and stick with them—not for a year or two, like most mentoring initiatives, but until they graduate.

The good news is how well this approach works. As I crisscrossed the country, seeking initiatives with the power to change kids’ lives, Friends of the Children is the only program I found that breaks the generational cycle of poverty, crime and teenage parenthood, enabling kids who begin with the odds heavily stacked against them to become engaged and productive citizens.

Coming of age can be a fraught journey, especially when there is no teacher or coach, no neighbor or clergyman in a youngster’s life—someone who listens, who can give a nudge at just the right moment, who can pry open the right door or maybe shut the wrong one. Ask anyone who has engaged with youngsters what matters most, and the answer is invariably the same—the presence of a caring and stable adult in a child’s life.

* * *

Those with faith in the transformative power of mentoring have always had a pocketful of feel-good tales about youngsters whose futures were reinvented by a solicitous adult. These stories may have motivated private benefactors to open their wallets, but they didn’t persuade policy-makers to do the same. In our results-driven era, the message is, “Don’t just tug at my heartstrings; prove that your social policy accomplishes something,” says Gary Walker, former president of Public/Private Ventures. Hard evidence that mentoring can make a difference came with a 1995 random impact study of Big Brothers Big Sisters, carried out by Walker’s organization, which found that after students were linked to a mentor for a year, school absenteeism was halved, first-time alcohol use was reduced by almost a third and first-time drug use was nearly halved. Suddenly a goody-two-shoes venture turned into a solid public investment.

The United States is a nation of joiners, and there are thousands of mentoring organizations in this country. Although these groups have the best intentions, they sometimes fail to honor their commitments. The consequences are unfortunate: mentoring relationships that fall apart in less than six months can actually do kids harm. To secure positive results, the Big Brothers Big Sisters study showed, it’s essential to maintain the consistent involvement of an adult in a youngster’s life for several hours each week, week in and week out. Ample time is needed, writes psychologist Urie Bronfenbrenner, to form “a special bond of mutual commitment” and an “emotional character of respect, loyalty, and identification.”

It’s this bond that Friends of the Children is cementing. The program began in 1993 in Portland, Oregon, and now has affiliates in six cities from Boston to Seattle. It’s the dream of Duncan Campbell, a 2009 winner of the prestigious Purpose Prize, and its inspiration comes straight from his life story. Campbell’s parents were alcoholics, and he has memories of long nights in the local bars, watching his mother and father toss down shots. “There were police and bill collectors at the house all the time. My father was sent to prison twice. We never had a conversation.” Campbell vowed that if he made enough money he’d do what he could to make sure that kids like himself got a better deal. After earning a fortune in timber investment, he set out to determine just what kind of help mattered the most.

The model relies on paid mentors, known as Friends, the majority of whom have worked with kids in some capacity in the past—as teachers, social workers, coaches or parents. They have embraced a job with modest pay, great stress, long hours—and considerable emotional reward. Every year they spend six weeks in the kindergarten classrooms of elementary schools in distressed neighborhoods, homing in on kids whose backgrounds might involve poverty, homelessness, domestic violence, and drug and alcohol abuse.

Each child is paired with a Friend, and the two of them get together every week for four hours. The bond between the child and the organization stretches to high school graduation. Many of these kids have grown hardened to adults who suddenly and unaccountably vanish, but because staffers typically remain with the organization for four or five years, transitions are smooth, sparing the youngsters the trauma of the revolving door. Together, youth and mentor explore their community or else spend time at the Friends House, a clubhouse and a refuge. Because children who read below grade level in third grade have only a one in seven chance of ever reading at grade level, the Friends and their young charges do a lot of reading together. They find their math puzzles on the streets of the city. “What’s the price of two slices of pizza? How much will it cost us to ride on the bus?”

By second grade, the children and their Friends are starting to do things with others in the group. Adolescents often go on group outings, since at this point in their lives they’re more interested in hanging out with their buddies than spending the day with adults, no matter how cool the grown-ups might be.

Throughout the relationship the mentors’ responsibility isn’t simply to boost their charges’ reading and math scores or to get them into good schools or make sure their cavities get filled. It’s doing whatever it takes to build up these kids’ social capital, making sure they’re ready for college or good jobs and a responsible, care-for-others life—doing whatever a knowledgeable uncle or godmother would do.

* * *

The New York City affiliate in central Harlem opened its doors in 2001. Central Park is only three blocks away, but so constricted are the horizons of the families that live here that some of the 5-year-olds had never set foot in the park. Executive Director Bob Houck dreams of a Friends House that fills an entire brownstone, but for now there’s a cramped, if comfortably furnished, basement apartment in a rundown building. The kids can do their homework there, work on the computer, try their hand at baking cupcakes, grow tomatoes on a sliver of earth or shoot hoops on the cement court. No city in the world offers more enticements for youth than New York, and the program takes full advantage of those opportunities, with outings to the Museum of Natural History, Coney Island, the Staten Island ferry, Yankees games and the ballet. On these outings, the Friends are on the lookout for anything that evokes a spark. “I bargained with Joey,” says Pedro Resto, one of the staffers. “Let’s bike across Central Park, go to the museum and then get a pretzel afterward.” Resto, who grew up just a few blocks from the Friends House, is a prize-winning filmmaker who became a Friend because “I want to give something back.”

Youth-serving programs that report great results sometimes cherry-pick the brightest youngsters or those who come from solid families. Friends of the Children does its cherry-picking in reverse, tackling the toughest cases. Among the youngsters in the New York program, 61 percent come from single-parent homes and 27.5 percent are in foster care or have no parent. Nearly half have a parent who is or has been in jail or prison, 25 percent were exposed to substance abuse, 18 percent have been abused and 20 percent are in special education classes.

The staff regularly gets together to share their highs and lows, and on the day I visit, Resto’s “low” is worrying. A second grader I’ll call Ramon has been a handful from the outset, and things are coming to a head. Four times since the start of the school year he has threatened to commit suicide. In the classroom he punches himself in the face, and he’s terrifying the other kids. “Your boy has Satan in him,” a neighbor told Ramon’s mother, who is raising her son on her own. She won’t consider medication, for she’s been told that it will turn Ramon into a zombie. Resto will try again to persuade Ramon’s mother that he should see a doctor; and he’ll push the school, which has laid all the responsibility on her, to devise an individualized education plan, as the law mandates. “This isn’t the toughest case I’ve had,” Resto reminds his colleagues. “I had a kid with bipolar disorder who was hospitalized for forty-five days and a second grader who threatened a girl with a knife. His dad had come home from prison, but then he went back. You’ve got to care for these kids, try to stay strong, but it’s devastating.”

New York’s statistics back up the success stories. Ninety-eight percent of the Friends students have been promoted every year (the only exceptions are a youngster who transferred to a charter school, where he was asked to repeat a grade, and a girl who lost a month because of family turmoil). The attendance record is 96.1 percent, better than the 91 percent citywide average; and the youngsters, many of them placed in the city’s best schools, have flourished. During the 2008-2009 school year, for the fifth straight time, their reading and math test scores were better than average in the schools they attended. All but one had stayed out of the juvenile justice system, and the only girl who had a child decided to put her baby up for adoption and stayed in school.

* * *

Friends of the Children’s progress in its home city of Portland, where the youth come from backgrounds similar to those of New York City youngsters, has been tracked the longest. The effects of the program there are mind-blowing: 85 percent of the kids have earned high school diplomas or GEDs—that’s at least 10 percent higher than the national average—and 60 percent have at least one parent who did not graduate. Forty percent go to college, also bettering the national average. While 60 percent have at least one parent who has been incarcerated, 95 percent stay out of the juvenile justice system. The best marker for teen pregnancy is being the child of a teen mother, but while 60 percent of these kids were born to a teen parent, 98 percent of the youth have avoided teen parenthood.

Last spring, the Harvard Business School Association of Oregon translated the Portland figures into the language of economics. The report concludes that because of the program’s impact on youngsters’ educational achievement, crime rates and teen parenthood, every dollar spent generates a greater than sixfold return. Six to one: that is light-years better than Head Start or Job Corps. Its impact on children is even more impressive than most of the renowned social policy experiments of the past forty years, which have prompted policy-makers to take kids’ concerns seriously. The National Institutes of Health is sponsoring a five-year evaluation of Friends of the Children, the first study of a long-term paid professional mentoring program. It’s too much to expect that the results will echo the initial findings, but given the long odds against these young people, Friends of the Children would be doing remarkably well, the return on investment still strikingly good, if just half of the kids’ lives were turned around.

National Executive Director Judy Stavisky hopes that the NIH study will prompt foundations and government agencies to take serious notice of the nonprofit. She knows the mentoring world well, since she previously had a large portfolio of mentoring programs at the Robert Wood Johnson Foundation; her connection is also intimate and personal, since for seventeen years she has mentored a young man who has recently started college. “Friends of the Children is a jewel,” she says, “and its promise is second to none.”

Bob Houck would like to see a Friends Place in every borough of New York City, a thousand children potentially saved from drowning. Duncan Campbell dreams bigger. “My vision is that every child in the country who needs a Friend would have one,” he says. But Friends of the Children currently enrolls just 700 youth in six cities, a teaspoonful in the ocean of need. The best estimate—prepared by MENTOR, a national nonprofit that oversees the field—is that 14.6 million youth aged 10 to 18 who most need mentoring lack the opportunity. Although other adults touch the lives of an uncounted number of youngsters, the estimate confirms the enormous gap between the existing pool of mentors and the potential need.

There will never be a sufficient number of volunteers; and volunteers can’t fill the gaps caused by inadequate medical care, bad schools and the like. As Marc Freedman, the guru in this field, argues, the mentoring movement “highlights an unmet need, goes part of the way toward redressing it, and calls out for reinforcements.”