Mariah Kelley was 14 credits short, weeks from giving birth, and seemingly light-years from getting a high school diploma when she was referred to the Performance Learning Center, which is located in a gritty industrial section of Richmond, Virginia.

A public-private partnership between the Richmond Public Schools and the Richmond affiliate of Communities in Schools, the Performance Learning Center takes the longest of long shots and, through individually tailored instructions, gets them to pay off — again and again.

The average GPA of an entering student is zero-point-nine. The percentage of students who leave with a high-school diploma is 96. A third go on to post-secondary education. “We know that kids can change their lives if given the opportunity,” says Wes Hamner, the school’s academic coordinator. The problem isn’t innate ability; it’s taking kids who derailed somewhere along the way and getting them back on the track to success.

Entire library floors groan under the weight of proposals for improving the schools. But education reform confronts a dilemma that better schools alone can’t fix: The best instruction in the world doesn’t do a bit of good to a kid who is standing on the corner instead of sitting in class. Nationwide, 7,000 students a day drop out. Nearly half are minorities. And without a diploma, they’re looking at lives likely to be one long, hard slog.

But students don’t simply wake up one morning and quit. They start to go off the rails long beforehand. Education is supposed to give everyone an equal start in life — but Harold Fitrer, president of the Richmond affiliate of Communities in Schools (CIS for short), says many pupils already have fallen behind before the first day of kindergarten: They haven’t been taught “numbers, letters, and colors.” Many then can’t make the crucial third-grade shift “from learning to read to reading to learn.” So they get held back. And get held back again. Before you know it, you have a 15-year-old seventh-grader who is frustrated, embarrassed, and dejected, with just enough math to figure out there’s no way he’ll ever graduate. He might drop out before entering high school — and never even get counted in graduation-rate statistics.

The problem is especially acute in Richmond, where 75 percent of children in the public schools come from families with incomes below the poverty line. In the Creighton Court housing project, in the city’s East End, adults who have a high school diploma are in the minority. Unemployment is rampant. Nearby Woodville Elementary School hosts holiday dinners because, says interim principal Kara Lancaster-Gay, “those don’t happen at home.”

School lockdowns are not mere drills. When Lancaster-Gay attended a vigil for the mother of one Woodville pupil, another pupil asked her why she had bothered: “Don’t you know people over here die every day?” A year and a half ago, Woodville staff were looking forward to watching the children enjoy a new $80,000 playground. Two days before classes began, somebody burned it down.

CIS plays a particularly active role in schools such as Woodville. An on-site services coordinator works with more than 20 partner groups, from the Junior League to the YMCA, to get kids the help they need. Help can be as simple as a free eye screening and a pair of eyeglasses, or as complex as grief counseling and intensive behavioral therapy for a pupil who has lost a parent to the prison system — or the cemetery.

Often the help comes in the form of food, including “pre-breakfast” — which is served to those kids who may not have eaten since lunch yesterday and are so hungry they need something even before breakfast is served. After school, dozens of children stay at the Boys and Girls Club until late evening, and get a good meal in the process. On Friday, CIS sends children home with a backpack filled with six meals from the Central Virginia Food Bank to tide them over until Monday.

But CIS does more than provide handouts. It also sets expectations surrounding the ABCs — attendance, behavior and coursework. CIS says 69 percent of its hardest cases improved attendance, 71 percent improved behavior and 75 percent improved their grades.

That’s what CIS claims, anyway. But you don’t have to take the group’s word for it. In 2011, the consulting firm ICF International completed a five-year study of CIS. The exhaustive examination compared 602 CIS schools with 602 demographically similar non-CIS schools. Just for good measure, it included controlled trials involving 573 students in two different states.

As Education Week reported, the study found “the program has a strong effect on reducing dropout rates and yielded small but consistent improvements in performance on state assessments for math. Results for reading and language arts tests were mixed. … Overall, the study concludes, the more fully and carefully the Communities in Schools model is put into practice, the more effective it is.”

None of this is cheap: Richmond’s CIS has a budget of more than $3 million, and nearly 1,800 volunteers give their time and expertise to the effort. And that goes on top of Richmond’s already-generous spending of nearly $14,000 per pupil, well above both the statewide average of $10,000 and the levels of nearby Henrico and Chesterfield, which also are about $10,000 per pupil. But spread across the 14,000 students who get CIS help, that $3 million comes to just a little more than $200 per student — which could make CIS one of the most cost-effective social interventions around.

There are 200 CIS affiliates around the country — five of them in Virginia, the rest scattered among 26 other states and the District of Columbia. All told, they help more than 1 million children every year. That kind of volume almost demands a cookie-cutter approach. But CIS may be effective precisely because it eschews uniformity. Although every CIS program abides by a few general principles, there is no procrustean, top-down formula dictated by a tower of bureaucrats in Washington or New York. Not only is every program different — every program treats every student differently. Those involved say that individualized attention is what makes it work.

When educators in Richmond and similar urban areas catch heck for lousy test scores, they often complain that they are being held responsible for factors they didn’t cause and can’t control. But that doesn’t mean those factors cannot be counteracted, or that they can be counteracted only by a massive infusion of government money and control. Private efforts can make a difference in public education. Just ask Mariah Kelley. After getting her diploma through the Performance Learning Center, she is now pursuing a degree in nurse anesthesia.

This article originally appeared at the Richmond Times-Dispatch.