As American schools struggle with issues of race, diversity and achievement, a new study in the American Sociological Review has split the difference in the ongoing discussion of resegregation（恢复种族隔离）. Yes, black, white and Hispanic students were less likely to share classrooms in 2010 than in 1993, but no, that increase in segregation is usually not the result of waning（渐亏的） efforts to reduce it. "People have a general idea that at the national level, there is widespread resegregation, based on the minority-white composition of the average school, says author Jeremy Fiel, a sociology Ph.D. candidate at UW-Madison.

A significant part of the reduction in classroom diversity is simply a result of the increasing share of the Hispanic population and the declining share of whites, Fiel says. "Blacks and Hispanics have attended schools with a smaller proportion of whites over time, but the composition of schools depends on the composition of the area.

"If an area is 50-50 black and white, like some metropolitan areas and non-metropolitan counties, you can't do anything to make the average black student's school more than 50 percent white or less than 50 percent black," he says.

Segregation is back in the news because many school districts are moving away from mandatory desegregation plans, and often these districts do become more segregated in terms of the balance of whites and minorities across schools, Fiel says.

But that finding, by itself, does not prove that intentional or reversible resegregation is taking place nationwide.

To study the causes of the increasing segregation of American schools, Fiel used data from the National Center for Education Statistics. Focusing on 1993-2010, he compared the racial makeup of schools to that of their surrounding areas, and calculated how school composition would look if all schools were desegregated to match local populations. "The difference between the actual change in school composition and the change in the hypothetical desegregated world is due to changes in policies that promote or reduce segregation," Fiel says.

Viewing it this way -- as a comparison of ideal to actual -- brought Fiel to a surprising conclusion. Even though minorities are attending schools with fewer whites, "the exposure of blacks and Hispanics to whites was actually higher than would be expected," given a massive change in the composition of the student population. "That's the major finding."