The Post focused on the student body within school districts, whereas other studies focus on students’ presence on a metro or statewide basis. The Post analysis measured integration only for districts that had enough racial diversity to allow for integration — what we call diverse districts.
In addition, this analysis uses a measure of segregation that accounts for demographic change in each school system, whereas others do not.
Q: Why was it important to look outside of big cities? Aren’t most students in large metro areas?
A: Schools in most big city districts, and many big suburbs, have long been segregated. The Post analysis finds that segregation often persists. But it also found that a surge of Latinos moving into smaller communities has translated into a big jump in the number of diverse school districts, and it found that those newly diverse districts were far more integrated than the big urban districts. These newly diverse districts are typically small on their own, but they add up to millions more children being educated in integrated schools than in 1995, the first year for which national data is available.
Some other research has also found high levels of integration in small towns and rural areas, such as a 2008 report from the Civil Rights Project at the University of California at Los Angeles.
Q: Why did The Post look only at integration within a school district? Aren’t some school systems overwhelmingly one race, with neighboring districts overwhelmingly another?
Yes, there are wide racial differences among school systems. Schools in Birmingham, Ala., are 92 percent black, for instance, while neighboring Mountain Brook schools are 96 percent white. But districts assign students to schools, for the most part, within the boundaries of their own systems. And in 1974, the Supreme Court ruled that, in most cases, courts cannot force suburbs to be part of a nearby city’s desegregation plan.
Q: What are diverse districts, and why did The Post focus on them?
A: Experts told us that a school district where one racial group makes up more than 75 percent of the students cannot realistically integrate itself, because there are not enough children in the minority. So we defined diverse districts as those where no one group makes up more than 75 percent of students. We found there was a large increase in the number of diverse districts between 1995 and 2007.
Q: If they are diverse, does that mean they are integrated?
A: Not necessarily. We considered districts to be integrated if they spread their diversity among the schools inside the district. So a district where all the black students are in one school and all the white students are in another would be considered segregated, whereas a district that had racial diversity in all its schools would be integrated.
Q: Do you consider schools that include black and Latino students to be integrated?
A: No. Both groups have been historically discriminated against, and experts say educating them together does not provide the same benefits as mixing them with white and Asian students, who tend to come from wealthier families and bring more resources. So our measure of segregation combines black and Latino students and looks at how well they are integrated with students of other races.
Q: Is it possible that you are missing a large number of segregated schools by excluding districts that are not diverse from the analysis?
A: The number of districts that are not diverse fell dramatically between 1995 and 2017, so this is not likely to be much of a factor. To be sure, we checked what would happen if we combined the non-diverse districts with the segregated districts. The results did not change.
Q: How did you quantify segregation and integration?
A: We used a measure called the variance ratio, sometimes called a correlation ratio, which looks at how isolated a racial group is, given the demographics of the district. We decided on this measure after considering two other approaches.
One of those approaches — exposure — measures the likelihood of interaction among students of different races, but it does not take into account the overall demographic makeup of a school district. So a school that is 75 percent black could receive a low exposure score even if the larger district is 75 percent black. This is important because whites make up a smaller share of the student population than once was the case, so it is not surprising when students of color have less exposure to white students than they used to.
We also considered measures of evenness — such as the dissimilarity index and entropy — which assess how closely each school matches the demographics of the district. One problem with dissimilarity is that it compares only two races at a time. Entropy includes multiple races, but it treats all races equally. So a district would get credit toward integration by mixing, for instance, Asian and white students, when it is black and Latino students who have been historically segregated.
The variance ratio contains elements of both exposure and evenness, because it’s a type of exposure measure that controls for the overall demographics of the district.
Q: You found integration on the rise. Is it possible that this will be fleeting? Will white families move out of communities that have attracted more children of color?
A: We don’t know. It’s possible that white students will flee these newly integrated districts. We found a small number of districts that became less diverse from 1995 to 2017. Most of these districts became predominantly Hispanic or black. The UCLA Civil Rights Project, using a different measure for segregation, found that integration rose in large suburbs only to fall again. The story could be different in the newly diverse, integrated districts identified by The Post. For instance, they are generally in smaller towns, where families have fewer educational and housing options.
For more details on The Post’s methodology, click here.