Glossary

Segregation refers to the extent to which students or residents of different groups (e.g. racial/ethnic groups, or economic groups) attend different schools or live in different neighborhoods. Segregation is a characteristic of a group of schools or neighborhoods (rather than a feature of any individual school or neighborhood). Segregation is at its minimum value if all schools/neighborhoods have the same racial/ethnic or economic composition, and is maximized if no individual is enrolled in a school or lives in a neighborhood with any member of any different racial/ethnic or economic group (i.e., if all schools/neighborhoods are racially or economically homogeneous). Typically, segregation is measured on a scale from 0 to 1, with 0 indicating no segregation and 1 indicating complete segregation.

School segregation estimates rely on racial/ethnic categories defined by the National Center for Education Statistics (NCES) Common Core of Data (CCD). White, Black, Asian, Native American, and Multiracial (identified since 2010-11 for all schools) refer to non-Hispanic students. Asian includes Asian, Hawaiian, and Pacific Islander students.

Residential segregation estimates use categories defined by the Census, and we use the same racial/ethnic categories as in the CCD. In 1970, racial groups are not defined with respect to Hispanic ethnicity; e.g., White residents include both those that identify as non-Hispanic and Hispanic. Multiracial individuals have been identified since 2000.

Free or reduced-price-lunch-eligible students.

Most traditional public-school districts in the U.S. are defined by a geographic catchment area. Some schools—chiefly charter schools and Bureau of Indian Education (BIE) schools—are located in a school district’s geographic boundaries even though they are not administered by that school district. We assign all schools—including charter and BIE schools—to geographic districts starting in 1998-99 (the year that charter and magnet indicators became available) by spatially joining latitude and longitude coordinates of schools’ physical locations to annual school district boundary shape files provided by NCES Education Demographic and Geographic Estimates (EDGE). Latitude and longitude data are not available prior to 2000-01; we assign schools in 1998-99 and 1999-2000 to the same geographic district to which they are geocoded starting in 2000. Some schools in the 1998 or 1999 data did not exist after 2000; we geocode these schools’ street locations (provided by CCD) to obtain latitude and longitude and then spatially join them to school districts. Schools can be spatially joined to unified, elementary, and/or secondary school districts; for non-unified districts, we use high- and low-grade data to assign schools to either elementary or secondary districts. We link neighborhoods to geographic school districts using geographic crosswalks that use population weighting to assign tracts’ populations to the school district(s) in which they are located. Segregation data in the Segregation Explorer is reported for geographic school districts.

Data in the Segregation Explorer is reported for geographic school districts. Data for administrative school districts can be downloaded on the Get the Data page.

Administrative school districts are defined per National Center for Education Statistics (NCES), and the schools operated by each administrative district are identified using the NCES district ID (leaid). Charter schools administered by a traditional (geographically defined) school district are including the administrative districts, but charter schools operated and administered by other organizations are reported as their own administrative districts.

Data in the Segregation Explorer is reported for geographic school districts. Data for administrative school districts can be downloaded on the Get the Data page.

We define neighborhoods as Census tracts. Census tracts are small geographic areas defined by the US Census Bureau; each typically has 1200-8000 residents.
The Explorer displays segregation for metropolitan areas defined as core based statistical areas (CBSAs) defined by the Office of Management and Budget (OMB) in 2013. The OMB divides the largest CBSAs (those containing a single Urban Area with a population of at least 2.5 million residents) into metropolitan divisions; we treat each metropolitan division as separate metropolitan area in the data files, rather than including their CBSAs as a single geographic entity. We also estimate segregation for CBSAs and metropolitan divisions using their boundaries as defined in 2003 and 2023; those estimates are available on the Get the Data page (along with other geographies not displayed, like commuting zones).

Understanding the Data

There are many ways to measure segregation. The primary index of segregation that we show in the Explorer is the Normalized Exposure Index (also known as the variance ratio index or the relative diversity index). It represents the difference in the composition of the average schools or neighborhoods of two population subgroups. For example, White-Black school segregation is defined as the difference between the proportion of White students in the average White student’s school and the proportion of White students in the average Black student’s school. If all schools have the same proportion of White students, then this difference will be 0, reflecting no segregation. If all White students attend schools that are 100% White, and all Black students attend schools that are 0% White, then this difference will be 1, indicating complete segregation. Other measures of segregation can be downloaded on the Get the Data page.

School segregation estimates draw on school-level enrollment data from the Longitudinal Imputed School Dataset (LISD) 1.0, created by the Segregation Tracking Project team. The LISD combines, cleans, and imputes missing and erroneous data from the National Center for Education Statistics Common Core of Data (CCD) since 1991. More information about the LISD is available at the Get the Data page.

Residential segregation estimates are derived from Census tract-level data on residents from the U.S. Census Bureau’s 1970 to 2020 decennial censuses and American Community Survey (ACS) five-year aggregate estimates from 2005-2009 through 2018-2022. The ACS trades annual frequency for smaller samples, so tract-level estimates require five-year aggregations.

Measures of segregation between other pairs of groups can be downloaded on the Get the Data page

To learn more about school segregation, see additional resources from The Unfinished Legacy of Brown v Board of Education at 70.

Data and resources on neighborhood segregation will be added to the Segregation Explorer soon.

All public schools—traditional, charter, and magnet schools—are included in the school segregation data, excluding special education, vocational, alternative, and virtual schools and those exclusively for deaf and blind students. All neighborhoods (census tracts) in the US are included in the neighborhood segregation data (the entire U.S. was divided into tracts for the first time in 1990; prior estimates are limited to tracted areas (metropolitan areas).
The school segregation data are based on enrollment data for grades K-12. For each school district, county, metropolitan area or other geographic unit, we measure segregation in each grade (among all schools that have that grade) and then average the grade-specific segregation measures across all grades, weighted by the number of students in each grade, to obtain an overall segregation measure for the geographic unit.
The Explorer shows between-school segregation in geographic school districts, counties, metropolitan areas, and states. Between-district segregation is shown for counties, metropolitan areas, and states. The Explorer also shows between-neighborhood residential segregation in school districts, counties, metropolitan areas, and states. The Explorer displays Normalized Exposure Index as the measure of segregation. To access data for other geographic units (e.g., administrative districts or commuting zones) or other measures of segregation (e.g., Dissimilarity Index and exposure indices) or segregation between different racial/ethnic or economic groups, please visit our Get the Data page.
Yes. These data are available for researchers to use under the terms of our data use agreement. You can download the data files and full technical documentation on our Get the Data page.
For questions about the Segregation Explorer and the data, or to report errata, please contact segxsupport@stanford.edu.