Glossary

The average test score indicates how well the average student in a school, district, or county performs on standardized tests. Importantly, many factors—both early in life and when children are in school—affect test performance. As a result, the average test scores in a school, district, or county reflect the total set of educational opportunities children have had from birth through middle school, including opportunities at home, in child-care and preschool programs, and among peers. Average test scores therefore reflect a mix of school quality and out-of-school educational opportunities.
As described in the Methods section, the test-score data in SEDA are adjusted so that a “4” represents the average test scores of 4th graders in 2019 nationally, a “5” represents the average test scores of 5th graders nationally, and so on. One “grade level” thus corresponds to the average per-grade increase in test scores for students nationally.
The average, or “mean,” is used to represent the typical or central value in a set of numbers.
The standard deviation indicates how spread out a set of numbers is relative to the average. The standard deviation represents the typical amount by which any single number differs from the average.
When numbers in a set have been “standardized,” they have been transformed so that the average value is exactly 0 and the standard deviation is exactly 1. After being standardized, positive numbers represent values that are above the average and negative numbers represent values that are below the average.
A standardized test used to measure the knowledge and skills a student has attained in a particular subject, after instruction.
The numeric score earned on an achievement test.
Achievement levels are test scores that have been collapsed into a small number of categories, each defined by a set of prespecified standards for what students should know and be able to do. Each achievement test has three to five achievement levels defined. Achievement levels often have labels describing performance relative to grade-level expectations, such as “basic,” “proficient,” or “advanced.”
The test scores, or boundaries, that separate a set of achievement levels.
A data set containing information about how many students in each group (where groups could be schools, districts, states, etc.) earned scores in each achievement level by year, grade, and subject.
The percentage of students whose test scores were at or above the “proficient” achievement level in each grade and subject.
A set of statistical methods used to facilitate comparisons across different tests.
The National Assessment of Educational Progress; see https://nces.ed.gov/nationsreportcard/about/ for more detail.

The National Assessment of Educational Progress; see https://nces.ed.gov/nationsreportcard/about/ for more detail.

A database of school proficiency data from state accountability testing housed by the National Center for Education Statistics; see https://edfacts.ed.gov/ for more detail.

The heteroskedastic ordered probit (HETOP) model is a statistical technique that can be used to analyze proficiency data.
Economically disadvantaged students.
Free or reduced-price-lunch-eligible students.

Understanding the Data

The data are based on the achievement tests in math and Reading Language Arts (RLA) administered annually by each state to all public-school students in grades 3–8 from 2018–19 to 2022-23. In these years, 3rd through 8th graders in U.S. public schools took roughly 90 million standardized math and RLA tests. We use 2019 data reported by EDFacts and 2021-22 and 2022-23 data reported publicly by states as the basis for the estimates shown in this explorer.

We combine information on the test scores in each administrative school district with information from the National Assessment of Educational Progress (NAEP; see https://nces.ed.gov/nationsreportcard/about/) to compare scores from state tests on a common national scale (see the Methods page).

We never see nor use individual test scores in this process. The raw data we receive includes only counts of students scoring at different test-score levels, not individual test scores. There is no individual or individually-identifiable information included in the raw or public data.

Educational opportunities include all experiences that help a child learn the skills assessed on achievement tests. These include opportunities both in early childhood and during the schooling years, and experiences in homes, in neighborhoods, in child-care and preschool programs, with peers, and in schools.

During the COVID-19 pandemic, students generally had access to fewer educational opportunities because of school closures, remote instruction, restricted engagement with peers and others outside of the household, and perhaps the reduced ability of their parents to provide educational support due to stress caused by the economic, social, and health impacts of the pandemic.

Average test scores reflect the cumulative educational opportunities available to students in a community. These opportunities may occur both in and outside of formal schooling.

Differences in average test scores over time reflect differences in the cumulative educational opportunities available to the students over time.

In the Recovery Explorer, we focus on two differences: (1) pre-COVID (2019) to one year post-COVID (2022); and (2) one year post-COVID (2022) to two years post-COVID (2023). The first tells us about how much educational opportunities were affected by COVID. Declines in test scores reflect declines in the cumulative opportunities to learn that students had before vs. during the pandemic. The second tells us about cumulative opportunities during the first full year of academic recovery post-COVID. Changes during this time tell us about whether students have been able to access more opportunities to learn (both in and out of school) during the 2022-23 school year.

We construct average 2019, 2022, and 2023 math (or RLA) test score estimates using state standardized math (or RLA) tests taken by public-school students in grades 3 through 8 in each year, and data from the NAEP. The change estimates in each subject are calculated as the difference in the average test score estimates between the two years. For example, the 2022-2023 change is calculated as the average math or reading test score in 2023 minus that in 2022.
The free/reduced-price lunch percentage measures the proportion of students in the school who are eligible for free or reduced-price lunches through the National School Lunch Program. Students are eligible for free or reduced-price lunches if their family income is below 185% of the poverty threshold. A school with a free lunch rate of 0% has no poor or near-poor students; the higher the free lunch rate, the greater the number of poor students. The lower the free/reduced-price lunch percentage, the more affluent the school. The data is obtained from the National Center for Education Statistics (NCES) Common Core of Data (CCD) database. We use multiple imputation to address two known issues in the free/reduced-price lunch data from the CCD: (1) the substantial missing data and (2) changes in reporting due to the community eligibility program in recent years. See the SEDA 4.1 technical documentation for more detail.
Percent of year in remote or hybrid learning is a measure of the percent of the 2020-21 school year that all students received remote and/or hybrid instruction, calculated from data courtesy of the American Enterprise Institute’s Return to Learn Tracker and COVID-19 School Data Hub.
Total ARP Elementary and Secondary School Emergency Relief (ESSER) funds are reported as a percent of the pre-pandemic district instructional budget, courtesy of Georgetown University’s Edunomics Lab. Instructional budgets obtained from the National Center for Education Statistics (NCES) Common Core of Data (CCD) database. ARP ESSER is represented; ESSER I and ESSER II allocations are not included.

There are several reasons why we may not show data for a particular state or district.

Estimates for all districts in a state may be missing because:

  • Sufficient data for estimation were not reported by the state in 2022 and/or 2023.
  • Fewer than 94% of students in the state participated in testing in the subject in 2019, 2022, or 2023.
  • The state changed its test or proficiency thresholds between 2022 and 2023.

Estimates for an individual district may be missing because:

  • Sufficient data for estimation were not reported for the district. For example, public data is often suppressed by the state to protect student privacy. When there is substantial suppression for an individual district, we may not be able to use the data to construct estimates.
  • The district is too small and/or has too few grades of data available to allow for the construction of reliable estimates.
  • The district does not have a geographic boundary; such districts include charter districts and/or specialized local education agencies.
  • Fewer than 94% of students in the district participated in testing in the subject in 2019.
  • More than 20% of students in the district took alternative assessments rather than the regular tests in 2019.

For more details, see the technical documentation.

Using the Opportunity Explorer

The Educational Opportunity Explorer offers three different ways of looking at educational opportunity in the U.S.: a map, a chart, and a “split screen" view of both. Use the Map | Chart | Map + Chart buttons in the header to select one of these views.

District data are available for most states; more states will be added as data become available. At the top of this Explorer, you can select from 6 Key Measures of educational opportunity to display in the map and chart:

  • 2019-2022 Change in Average Math and Reading Scores, which reflect the difference in educational opportunities and experiences between cohorts of students tested in grades 3-8 in 2019 and 2022.
  • 2022-2023 Change in Average Math and Reading Scores, which reflect the difference in educational opportunities and experiences between cohorts of students tested in grades 3-8 in 2022 and 2023.
  • 2019-2023 Change in Average Math and Reading Scores, which reflect the difference in educational opportunities and experiences between cohorts of students tested in grades 3-8 in 2019 and 2023.

Performance is measured in grade levels, relative to the national average in grades 3-8 in 2019. Just below these buttons, you can use the left-side Data Options panel to filter by demographic (e.g., “all students"), and the type of places (states, or school districts) you’d like to display. In the Chart view, you can choose to view the entire country, or individual states.

Light gray represents zero, or “no change” in average test scores.

Positive changes in average test scores are shown in green and negative changes in blue. Darker shaded circles represent larger changes.

Colorblind users: We have made efforts to ensure accessibility for the most common forms of colorblindness. For less common forms (such as tritanopia), colors may be less distinguishable; however, the data are still accessible in the map legend, charts, and other displays.

You can view administrative school district data for most states, as well as state-level estimates. To change between location types, use the Region menu in the Data Options panel at left.

You can navigate to your desired location via the navigation controls in the map. Or type a location name into the search bar in the header.

Hovering (or on a touch device, tapping) on a location in the map or chart will show an overview of that location’s data. Clicking or tapping on the location will open a Location Panel that shows a full view of all available data, as well as options for viewing other selected locations.

Clicking or tapping on locations will also add them to the Locations menu within the left-side Data Options panel. (You can add an unlimited number of locations.) Click or tap on any location’s tab to highlight it in the map or chart, and the Locations menu.

You can select an unlimited number of locations at once. Selected locations will be stored in the Locations menu within the left-side Data Options menu.
Clicking or tapping on any location will open a Location Panel that offers a full view of the data.

In the Data Options column displayed on the left, click Data Filters. Using the controls in this panel, you can filter what data is displayed on the map.

When multiple selections are made, the filters work together to narrow what is displayed. To return to the default view, click the “Reset data filters” button. If a filter is not available, you will not be able to select it.

The chart shows districts in the U.S. Each circle represents a district, and the circle’s size is proportionate to the number of students (e.g., a larger circle is a district with more students). On the vertical (Y) axis is the range of the selected metric in grade levels. The horizontal (X) axis shows one of three variables (you can select which you would like to show): the free and reduced lunch eligibility rate in the district, the percent of the school year spent remotely during the pandemic, or the ESSER funding received by the district.
This map can be customized to show the change in math and reading test scores between 2019 and 2022, 2022 and 2023, or 2019 and 2023 for students in U.S. districts or states. The test scores represented here were collected in grades 3-8 from 2018-19 and 2022-23 at public elementary and middle schools in the U.S.