Just beyond the bodegas painted in tropical hues, past the bleak jail for juveniles and the vacancy signs along Broad Street in Hartford, Connecticut, a startlingly sleek, sterile collection of buildings materializes. Weekday mornings, a chain of yellow buses encircles the compound. Under the eyes of security guards and cameras, kids hop down, saunter into buildings and settle into classrooms where the mix of complexions and family incomes does not match Census data culled from these streets.

Many of the children scattered among the elementary, middle and two high schools have indeed been “bused” in, to engineer the creation of racially and economically diverse schools in this otherwise extremely poor, sharply segregated Latino neighborhood. Some of the children who attend the schools in this “learning corridor” live nearby. Others come from the African-American neighborhoods to the north, and a large share travel up to an hour from the lily-white suburbs that surround the city of Hartford, where 46 percent of children are poor. Several other “magnet” schools in and around the city open their doors each morning to a student body that reflects the diversity of the region, as opposed to the homogeneity found in schools that enroll kids from just one town or neighborhood.

“It has been nothing short of a beautiful experience,” says Mara Whitman, a white mother of four who opted for a magnet in Hartford over the far more affluent and far less diverse schools in her town, West Hartford. “To be honest, it was not the diversity that attracted us. It was the educational program. The theories that drove instruction were well thought, based in evidence…. But it wasn’t long before we realized that the diversity made the experience rich.”

After a state court ruled in 1996 that the region’s public schools were segregated, in violation of Connecticut’s Constitution, magnet schools became the principal remedy. Urban and suburban parents alike were quick to sign their children up for the new schools, which were often (though not always) located in urban communities of color. The waiting lists grew long for the more successful schools. Thanks to an unyielding push by civil rights lawyers, state legislators went on to approve funding for more magnets. Now, in one of the nation’s most segregated and economically stratified regions, tiny centers of racial and economic integration have taken root. Significantly—like students attending magnet schools in California, Nevada, Florida, Illinois and other areas—children in Connecticut’s magnet schools have registered promising academic results, often outdoing those in traditional public schools, which were, by comparison, sharply segregated by race and class. This past December, a study in the peer-reviewed journal Educational Evaluation and Policy Analysis compared academic results between students who’d applied to Connecticut’s magnets and were not selected through the blind lottery and students who were selected and got to attend a magnet school. The magnet school students who lived in urban ZIP codes (these students are mostly Latino or black) made greater gains in math and reading than did their fellow students who stayed in the urban (segregated) schools. What’s more, the suburban students—this group is largely white—attending magnets outdid their peers at traditional suburban (and generally much whiter) schools, too. The “achievement gap” between white students and students of color tended to be smaller in Connecticut’s magnet schools than it was in traditional schools.

While research on magnets and academic achievement is far from conclusive and often complicated by methodological concerns, you might think that the increasingly promising results emerging from a variety of locations, coupled with magnets’ desegregation mission, would make them serious contenders for the $4.3 billion made available through the Education Department’s competitive grant program, Race to the Top. But the Obama administration’s reform strategy mostly overlooks the nation’s thousands of magnet schools. Instead, administration officials much more strongly favor a newer “reform”—charter schools, which demonstrate no evidence of sustained, large-scale success and tend to compound racial and economic segregation.

Charter schools are similar to magnet schools in that students choose to attend them, and the schools usually offer a specialized theme or curricular strength, such as visual arts or science. But magnets, begun in the 1960s and expanded in the ’80s as voluntary desegregation measures, are operated by public schools or publicly funded agencies that work closely with public schools. They are thus subject to all the regulations (and union rules and civil rights protections) associated with public schools. Though some magnets have clearly strayed from their original desegregation mission, the magnet schools that operate as originally intended either attract a racially diverse student body or are at least actively trying to. They usually provide free public transportation for students to travel longer than typical distances. Charter schools, on the other hand, can be operated by just about anyone, including private corporations, nonprofits, individuals who may or may not have experience with teaching children or operating an educational institution, or religious groups in partnership with other entities. They don’t necessarily need to offer free transportation, and they are under no mandates to be racially or economically diverse.

Charter schools have enjoyed enthusiastic rhetorical support since Obama’s presidential campaign and, in recent months, increasing monetary support. In March 2009, the president told the US Hispanic Chamber of Commerce: “I call on states to reform their charter rules and lift caps on the number of allowable charter schools wherever such caps are in place.” In public statements about Race to the Top, Education Secretary Arne Duncan made it clear that states that limit charter school expansion would not be looked upon favorably in Race to the Top’s competitive grant program.

Lawmakers in several states, strapped for cash mid-recession, either did or at least tried to do exactly what the administration told them to do. Some did it fast, passing laws that supported the development of more charters, without evidence that such schools were consistently doing well by children. In response to the administration’s pressure, Louisiana, Massachusetts, Illinois, Tennessee and Delaware either lifted their caps on charter schools or let their moratoriums lapse. And not surprisingly, the winners in the first round of funding, announced at the end of this past March, were Delaware and Tennessee, where officials had significantly loosened restrictions on how many charter schools could operate. In Tennessee, legislators had even created a new funding stream to pay for charter school facilities. Together, the two states will receive about $600 million, not only for charters, of course, but for larger reform plans in which charters play an important role.

The Education Department’s proposed 2011 budget would spend $490 million to promote school choice, the vast majority of which would go toward charter schools. This amounts to a 20 percent increase—or about $81 million—over what charters receive in the current budget. By comparison, federal funding for magnet schools comes mainly through the Magnet Schools Assistance Program. This year, MSAP will provide about $100 million to some forty schools across the country. The program, though, has been flat-funded since 2008. President Obama’s proposed 2011 budget does include a modest funding increase for the program, bringing it up to $110 million.

“The federal government is beginning to pay us a little bit more attention,” says Robert Brooks, executive director of the Washington-based Magnet Schools of America, which sponsors conferences and provides technical services to magnet schools. “But this overemphasis on charter schools is astonishing. Magnet schools have a much longer record of success in terms of academic achievement and equity. I think maybe the difference is that charter schools are viewed as a newer concept. People tend to like the idea of a newer concept. Legislators especially like to be associated with newer concepts.”

Magnets, Brooks supposes, are unfairly suffering from their roots as “equity-minded institutions from the ’70s that sought integration.” Now, he adds, “there are all kinds of people questioning whether that’s even a valid way to educate children anymore, which flies in the face of what the research says. But we are not straying from our message: diversity is beneficial for kids and for society.”

The charter/magnet imbalance has unnerved many advocates, who are urging federal officials to give the well-performing, racially diverse magnet schools consideration at least equal to charters. (Full disclosure: the Harvard-based Institute where I work is a member of the National Coalition on School Diversity, which has submitted formal comment letters to the administration on related issues.)

“We think that the practice of ignoring magnet school successes and pushing charters as the only major focus of choice in federal initiatives…ignores some strong evidence about [magnets’] benefits,” Professor Gary Orfield, co-director of UCLA’s Civil Rights Project, wrote in a 2009 memo to top Education Department officials. “In thinking about school choice and innovation we recommend that…increases in resources be competitively available between both options, and that civil rights policies common in magnets be extended to charters.”

The Civil Rights Project released a study in January demonstrating that charter schools tend to exacerbate segregation, which is linked to a host of short- and long-term educational and social inequalities. African-Americans, our most segregated racial group, are the students most segregated in charter schools. According to the report, 70 percent of black students enrolled in charter schools attend schools that are 90 to 100 percent nonwhite. This is about twice the share of black public school students who attend schools that are segregated. What’s more, the report found that charter schools failed to produce data related to enrollment and accessibility for students who are still learning English or about students from low-income families.

Populist commentators of all ideological stripes have won easy applause in recent years by passionately insisting that when it comes to learning, demographics do not matter. They point to isolated examples of all-black or all-Latino schools—often charter schools—that “beat the odds.” However, the notion that high-poverty racially segregated schools are equal to middle-class schools has been consistently disproved by a half-century of research. Loudly heralding the relative few that “beat the odds” (typically not for more than two years) obscures the harsh reality of the odds themselves, which remain low for kids who attend high-poverty segregated schools.

The weight of social science evidence demonstrates that racially diverse schools are associated with achievement in math and reading for Latino and African-American students, with more advanced critical thinking skills, intellectual engagement and a reduction in racial stereotyping. Learning challenges that may be related to stress, poverty, health problems or neighborhood violence, researchers are starting to understand, may be more effectively overcome in schools that are not sharply segregated by race and class.

The decades of findings on these points have led to something rare in social science—a consensus. In 2007 the National Academy of Education wrote, “In summary, the research evidence supports the conclusion that the overall academic and social effects of increased racial diversity are likely to be positive. Racial diversity per se does not guarantee such positive outcomes, but it provides the necessary conditions under which other educational policies can facilitate improved academic achievement, improved intergroup relations, and positive long-term outcomes.”

Meanwhile, there is no consistent evidence and certainly no research consensus that charter schools, just by virtue of being charter schools, do any better, on average, than traditional public schools. A 2009 study conducted in Boston returned typical results. Some charters there did better than the traditional public schools, and some did worse. A study of charters in Minnesota, which has the nation’s oldest system of charters, found that students in charters generally did less well on achievement tests than similar students in public schools. That study, conducted by the University of Minnesota–based Institute on Race and Poverty, also found that charter schools tended to worsen segregation levels in Minneapolis and St. Paul. Researchers concluded that “although a few charter schools perform well on standardized tests, most offer low income parents and parents of color an inferior choice—a choice between low-performing traditional public schools and charter schools that perform even worse.”

A 2005 study from the Economic Policy Institute found that in Florida, Michigan, Texas and Pennsylvania, students in charter schools that had been operating for more than three years did no better on state tests than public school students did. The number of charter schools, however, has more than doubled over the past decade, from about 2,300 in 2001 to nearly 5,000 now. According to Magnet Schools of America, there are about 4,000 magnet schools in the United States, enrolling about 2.5 million students.

That charters do not upset the stratified structure of public education—i.e., they tend to leave white kids in “white” schools and African-American students in identifiable “black” schools—may be exactly what makes them, at first glance, appear politically neater than magnet schools. After all, if magnets work as intended, they absolutely will alter the status quo of entrenched residential segregation by allowing kids of differing racial backgrounds to pursue opportunity across man-made boundaries. (As even magnet advocates are quick to point out, not all magnets are racially diverse. Those that are not, diversity advocates argue, should not be the ones that are rewarded with federal dollars.)

There is no sign that the administration might be open to rethinking its infatuation with charters. However, the mission of magnet schools seems to be in line with the administration’s recently and clearly articulated concerns about racial segregation and concentrated poverty—creating an opening for advocates to press for magnet school funding increases beyond what the administration has already proposed.

In March, Secretary Duncan stood at the Edmund Pettus Bridge in Selma, Alabama, and reiterated a statement Barack Obama had made in Philadelphia as a presidential candidate: “Segregated schools were, and are, inferior schools. We still haven’t fixed them fifty years after Brown v. Board of Education—and the inferior education they provided, then and now, helps explain the pervasive achievement gap between today’s black and white students.”

Similarly, at a Black History Month celebration recently, Housing and Urban Development Secretary Shaun Donovan stated even more clearly a commitment to reducing segregation and concentrated poverty. “The neighborhoods of concentrated poverty we see in communities across America didn’t result in spite of government—but in many cases because of it,” Donovan said, touting the administration’s “commitment to building more inclusive communities.” He added that “we expect communities receiving federal funding to end practices that limit diversity and start promoting stable, inclusive communities.”

Civil rights groups view magnet schools as a potentially good offering for an “inclusion”-seeking Democratic administration to carry across the legislative aisle. Magnets embody “choice,” and so should win support from conservatives who advocate free-market solutions for public schools. Magnet schools’ survival hinges on their knack for “innovation,” as educators must create curricular themes that attract families. As public schools, magnets are “accountable” and are judged on results from state-administered tests. Modern buzzwords apply, but at heart magnets remain civil rights–era remedies to structural inequality. Even if they aren’t shiny and new, magnets might come close to that rare mix of having something for everyone.

“I hear about other schools that are all-black or all-white trying to teach about how the society is diverse and trying to get it through to kids that they have to understand other perspectives,” says Lorna Shipp-Parmlee, who has three children attending magnets who ordinarily would have been assigned to high-poverty schools in their predominantly black Hartford neighborhood. “But when you are in a school that really is diverse, you aren’t learning diversity—you are actually living it.”