Achievement gaps are visible in both measured performance and access to educational opportunities. In the United States, 2022 NAEP results showed a 29-point mathematics gap between White and Black fourth-graders, while PISA 2022 results found a 93-point mathematics difference between socioeconomically advantaged and disadvantaged 15-year-olds across OECD countries. The measures below cover different populations, subjects, and assessment systems, so their points and percentages should not be treated as interchangeable.
Contents
- U.S. racial and ethnic gaps
- Reading performance across the distribution
- Access to advanced and accelerated courses
- International socioeconomic gaps in mathematics
- Reading, science, and high performance internationally
- How gaps changed across PISA cycles
U.S. racial and ethnic gaps
The 2022 National Assessment of Educational Progress (NAEP) measured substantial differences in fourth-grade mathematics scores. White fourth-graders scored 29 NAEP mathematics points higher than Black fourth-graders and 21 points higher than Hispanic fourth-graders. Fourth-grade boys scored 6 points higher than girls. These are assessment scale-point differences, not percentage-point differences. NAEP Mathematics: National Student Group Scores and Score Gaps
The longer civics trend shows that a gap can narrow without every group’s score rising. The White–Hispanic eighth-grade NAEP civics gap was 19 points in 2022, compared with 31 points in 1998. Hispanic eighth-graders’ score rose 14 points over that period, while the White score did not change significantly. NAEP Civics: Student Group Scores and Score Gaps
The pandemic-era trend for nine-year-olds moved in the opposite direction for one racial comparison: the White–Black NAEP mathematics gap widened by 8 points between 2020 and 2022. The same report found that reading and mathematics declines were greater at the 10th and 25th percentiles than among higher performers. That distributional pattern matters because an overall average can conceal larger losses among lower-performing students. Reading and mathematics scores for 9-year-olds decline during pandemic
Reading performance across the distribution
The 2024 NAEP reading results show how widely fourth-grade performance was spread. The average was 215 points, 2 points below the 2022 average. At the 50th percentile, the score was also 215 points; the 75th percentile was 245, and the 90th percentile was 265. At the lower end, the 25th-percentile score was 189 and the 10th-percentile score was 158. 2024 NAEP Mathematics and Reading, Grades 4 and 8 – Infographic
| Fourth-grade NAEP reading measure | 2024 score or share |
|---|---|
| 90th percentile | 265 points |
| 75th percentile | 245 points |
| 50th percentile | 215 points |
| 25th percentile | 189 points |
| 10th percentile | 158 points |
| Reached Proficient | 31% |
| Reached Basic | 60% |
| Reached Advanced | 8% |
Achievement levels provide another view of the same assessment. In 2024, 31% of fourth-graders reached NAEP Proficient in reading, 60% reached at least NAEP Basic, and 8% reached the NAEP Advanced level. “Basic,” “Proficient,” and “Advanced” are NAEP achievement-level categories; they are not the same as percentile ranks. NAEP Reading: National Trends and Student Skills
Together, the percentile results and achievement-level shares describe a broad performance distribution rather than a single national outcome. The 107-point distance between the 90th-percentile score and the 10th-percentile score is a descriptive spread within fourth-grade reading in 2024. It is not a racial, socioeconomic, or causal gap by itself.
Access to advanced and accelerated courses
Opportunity gaps can appear before an outcome is measured. The 2020–21 Civil Rights Data Collection compared student participation in advanced opportunities with overall high-school enrollment in U.S. public schools.
Latino students accounted for 27% of high-school enrollment but 20% of AP science and computer-science enrollment. They also represented 19% of AP mathematics enrollment. White students were 48% of high-school enrollment and 60% of dual-enrollment students, while Asian students were 5% of high-school enrollment and 6% of dual-enrollment students. About 1.6 million students enrolled in dual-enrollment or dual-credit programs during 2020–21. 2020-21 Civil Rights Data Collection: A First Look: Students’ Access to Educational Opportunities in U.S. Public Schools
| Group or measure | Overall enrollment | Opportunity enrollment |
|---|---|---|
| Latino students, high school vs. AP science and computer science | 27% | 20% |
| Latino students, high school vs. AP mathematics | 27% | 19% |
| White students, high school vs. dual enrollment | 48% | 60% |
| Asian students, high school vs. dual enrollment | 5% | 6% |
Older federal data show similar representation differences in other advanced programs, but the measurement period is earlier and should remain separate. In 2011–12, White and Asian students were 70% of gifted-and-talented enrollment but 55% of enrollment in schools offering those programs. Black and Latino students were 26% of gifted-and-talented enrollment but 40% of enrollment in offering schools. Protecting Civil Rights, Advancing Equity: FY 2013–14
The same 2011–12 source reported middle-school Algebra I pass rates of 86% for White students and 72% for Black and Latino students. For Algebra I taken in grades 11 or 12, the pass rate was 69% for English learners and 71% for non-English learners. These older course and pass-rate figures are not estimates of 2020–21 participation and are not directly comparable with the newer CRDC enrollment shares.
International socioeconomic gaps in mathematics
PISA 2022 assessed 15-year-olds and used the ESCS index to describe socioeconomic status. Across OECD countries, socioeconomically advantaged students scored 93 PISA mathematics points higher than disadvantaged students. Forty-seven percent of disadvantaged students scored below mathematics Level 2, compared with 14% of advantaged students—a 33-percentage-point low-performance gap. Disadvantaged students were 7 times more likely to score below Level 2. Equity in education in PISA 2022: PISA 2022 Results (Volume I)
Socioeconomic status explained 15% of within-country variation in mathematics performance in PISA 2022. Across OECD countries, one ESCS index unit corresponded to 39 additional mathematics points. The socioeconomic gradient was steepest in the Slovak Republic, at 53 mathematics points per ESCS unit. The Czech Republic, Israel, and Singapore each recorded 51 points per ESCS unit.
The size of the gap varied across education systems. The mathematics difference between students in the top and bottom socioeconomic quarters exceeded 93 points in 22 PISA 2022 countries or economies. It was 50 points or less in 13 countries or economies. These comparisons describe associations between socioeconomic status and performance; they do not by themselves establish that status caused a particular score.
Reading, science, and high performance internationally
The relationship between socioeconomic status and achievement extended beyond mathematics in PISA 2022. Socioeconomic status explained 13% of within-country variation in reading performance and 14% of within-country variation in science performance across the participating OECD context. Equity in education in PISA 2022: PISA 2022 Results (Volume I)
Disadvantaged students were more than 5 times as likely as advantaged students to be low performers in reading. They were also more than 5 times as likely to be low performers in science. These are relative-likelihood comparisons, not percentage-point differences. Executive Summary: PISA 2022 Results (Volume I)
Disadvantaged students were underrepresented among the highest performers. Across OECD countries, they made up 10% of the top quarter of national mathematics performers, 11% of the top quarter in reading, and 11% of the top quarter in science. The average within-country difference between students at the 90th and 10th ESCS percentiles was 2.34 index points; the difference exceeded 2.6 index points in 28 PISA 2022 countries or economies. Equity in education in PISA 2022: PISA 2022 Results (Volume I)
How gaps changed across PISA cycles
PISA trend data show that socioeconomic gaps do not move uniformly. Between 2018 and 2022, the mathematics socioeconomic gap did not change in 51 of 68 countries or economies. Across OECD countries, however, the gap widened by 7 points on average. It widened in 12 countries or economies and narrowed in 5. Changes in performance and equity in education between 2018 and 2022: PISA 2022 Results (Volume I)
The performance of disadvantaged students fell by 17 mathematics points on average across OECD countries from 2018 to 2022, and disadvantaged students’ performance declined in 34 PISA countries or economies. Albania recorded an almost 70-point decline for disadvantaged students. In the Philippines, the mathematics socioeconomic gap narrowed by 38 points over the same cycle. A narrower gap can therefore reflect different changes in the groups being compared, so the gap trend should be read alongside performance trends.
Over the longer period to PISA 2022, the mathematics socioeconomic gap stayed stable in 42 of 62 countries or economies. Across OECD countries, it widened by 3 points on average and narrowed in 12 countries or economies. The long-term comparison covers 2012 or 2015, depending on the country or economy, through 2022; it is not a single identical-length series for every participant. Long-term trends in performance and equity in education: PISA 2022 Results (Volume I)
NAEP and PISA use different populations, grades or ages, subjects, subgroup definitions, scales, and significance conventions. NAEP values above are assessment scale points; PISA values are PISA points; ESCS is an index. The U.S. course-access figures describe public-school participation, and the 2011–12 opportunity figures are explicitly older measurements. Those scope limits are essential when comparing achievement gaps across sources.