The blog of the Urban Institute
August 26, 2014

America's public schools remain highly segregated

August 27, 2014

Fifty million children will start school this week as historic changes are under way in the U.S. public school system. As of 2011 48 percent of all public school students were poor* and this year, students of color will account for the majority of public school students for the first time in US history.

What is surprising about these shifts is that they are not leading to more diverse schools. In fact, the Civil Rights Project has shown that black students are just as segregated today as they were in the late 1960s, when serious enforcement of desegregation plans first began following the passage of the 1964 Civil Rights Act.

Despite our country’s growing diversity, our public schools provide little contact between white students and students of color. We’ve mapped data about the racial composition of US public schools to shed light on today’s patterns at the county level. These maps show that America’s public schools are highly segregated by race and income, with the declining share of white students typically concentrated in schools with other white students and the growing share of Latino students concentrated into low-income public schools with other students of color.

In every state but New Mexico and Hawaii, the average white student attends a school that is majority white.This is unsurprising for large swaths of the Northwest, Great Plains, Upper Midwest, and Northeast, which are home to very few kids of color. But even in diverse states like Illinois, Michigan, Pennsylvania, and New York, few white children attend diverse schools.

The separation of races is most clearly seen in large metropolitan counties that hold the bulk of a state’s population and most of its students of color. For example, in Chicago (Cook County), the overall student population is about 25 percent white, 31 percent black, and 37 percent Latino, but 96 percent of black students attend majority non-white schools and 67 percent of white students attend majority white schools. In other words, white students tend to attend schools with other white students and black and Latino students attend schools with other students of color. Similar patterns emerge in other large midwestern cities like Detroit (Wayne County), Minneapolis-St. Paul (Ramsey and Hennepin County), and Indianapolis (Marion County).

In Worcester County, MA, white students account for 73 percent of the overall student body and almost all of them (91 percent) attend a majority white school, while the few Latino and black students in the county (15 and 5 percent, respectively) typically attend majority non-white schools.

In the South and Southwest, where the number of Latino students is growing especially fast, Latino students typically attend majority non-white schools with other children of color, while the few white children in these areas attend schools that are majority white.

In an increasingly diverse society, our public schools give us the unique opportunity to cross traditional racial and class boundaries. Ideally, they would be spaces where students can interact with and learn from peers with backgrounds different than their own, ensuring that future generations have friends outside their own racial group and helping mold them into productive members of a multi-racial society. Unfortunately, this potentially productive exchange is not happening in most public schools across the country.

We hope these maps provide a starting point for further analysis and for serious conversations—at local, state, and national levels—about the complex forces sustaining school segregation and the actions we need to take if we want our public schools to better reflect the diversity of our population.

The post originally stated that over half of all public school students were poor. It was 48 percent as of 2011.

Follow Reed Jordan on Twitter.


As an organization, the Urban Institute does not take positions on issues. Experts are independent and empowered to share their evidence-based views and recommendations shaped by research.


That’s a good point. If our public schools nationwide were perfectly integrated, then no white students would attend a majority-white school. Though as I discuss in the post, most students of color in the U.S. live in major metropolitan areas, and large swaths of the country have very few students of color. So we should be examining those metropolitan areas that have a lot of students of color to understand what school segregation looks like. Put another way, we should be asking: in areas with large percentages of students of color, do the white students attend schools with these students of color? Or are the white students attending schools with other white students? If you look at some of the major cities and metropolitan areas in the U.S. - such as Chicago (Cook County), Indianapolis (Marion County), the various counties of New York City, ect. –you see that they tend to be fairly diverse. But the white students in these areas tend to attend schools with other white students, and students of color attend schools with other students of color (a “majority non-white” school). This is what we can call school segregation. For example, only 25% of students in Chicago are white, but most of them (67%) attend a majority white school. If the Chicago schools were integrated, then no white students would be in a majority-white school.
This is just a purposefully inflammatory way of looking at a more fundamental set of data points. The VAST majority of students & families in this country do not get to arbitrarily choose the school they attend. They are assigned a school based on the location of their home. Thus, if the family lives in a community full of like-raced individuals, their children will all be attending the same school. So the question is NOT "Why are our schools segregated?" The question should be more basically asked " Why do people tend to live in communities of the same race?" That's a question that can't really be politically correctly asked.
The county where I grew up is totally misrepresented in this, and it makes me wonder how much of the rest of the country is misrepresented in the same way. The notion that the figures at the county level show that the schools are racially segregated is completely false. Each town in the county has only one school, so every child in town goes to the same school. All the demographics would show in this case is how many people of each race happen to live in each town. It's not just my county either, this would apply to the majority of the counties in the state, because very few counties in the state have any big cities in them and most towns have only one school. Rolling things up to the county level makes it look like segregation, only because you've arbitrarily chosen to look at counties, and because you're suggesting that the distribution of students without respect to the actual population equals segregation. It happens that the vast majority of latinos in the county live in one town, and there are very few in the other towns. If you blend together the figures for the whole county as you have, you've created nothing but a baseless inflammatory statistic. I realize large population centers are going to be more applicable to your study, but publishing a map depicting the entire country this way makes it look like you don't give much weight to that. More personally I'm a little bit offended at the implication of institutionalized racial segregation in the place where i grew up, when I know it isn't the case there.
Whites tend to live near other whites, Blacks tend to live near other Blacks, Latinos etc. Given this, it's no surprise that the local schools reflect this makeup in their racial mix. You cite Chicago as an example, but Chicago is a *huge* place with lots of local neighborhoods, more often than not racially homogenous, and a school or two in each.
I'm wondering if it might be helpful to only show those counties for which, as you mention, have large percentages of students of color with white students still attending majority white schools (or vice versa). Without additional filtering of the counties, these maps are difficult to interpret on the national level. (So I agree with WJ about needed a little more tweaking to really have the story clear without clicking on every county). Thanks for the analysis!
These maps imply a lot but don't really say anything. If we were to throw up a map of China and ask what percentage of Asians attend a predominately Asian school, the country would be pitch black. During Jim Crow, students were forceably segregated, which was wrong. That's completely different from voluntary segregation. Did anyone actually expect school busing to change the fundamental nature of people to surround themselves with people with similar backgrounds and life experiences?
Given the relative percentages in the country, I'm not sure that this proves anything. After all, if every single school reflected the racial make-up of the country exactly, there would be zero schools which didn't have a white majority. (At least for a few more years.) If you want to make a case for segregation in schools, why not graph how many children attend schools where the number of children of other races differs by more than, say, 10% from the percentages for the state as a whole? You would still, no doubt, find some, albeit fewer. But at least the data would be meaningful.
The statement that 48% of public school students are poor is not correct. 48% of public school students are low income, which includes poor plus what might be called near-poor. The figure is the percent of students who receive free or reduced price school lunches.