Federal Test Data Shows US Students Are Still Behind Pre-Pandemic Levels in Maths and Falling Further Behind China

America's 13-year-olds have made no measurable progress in maths or reading since 2023, with current scores remaining significantly lower than those recorded before the pandemic, according to new federal data released on 10 June 2026.

The National Assessment of Educational Progress long-term trend results show a stark generational divide. Nine-year-olds posted gains of four points in both maths and reading compared to 2022, a development federal officials attributed to this cohort entering school largely after pandemic-era closures had ended. Thirteen-year-olds, whose foundational learning years coincided directly with COVID-19 disruptions, recorded no measurable improvement in either subject since the last assessment and continue to trail their pre-pandemic predecessors in both disciplines.

Matthew Soldner, acting commissioner of the National Centre for Education Statistics, told the reporters that 'the story is very different for our nation's 13-year-olds.' He described the 13-year-old cohort's results as raising 'huge questions' that 'ought to serve as a catalyst for change.'

Thirteen-Year-Olds Stuck Below Pre-Pandemic Scores Since 2023

The NAEP long-term trend assessment has tracked American students in reading and maths since the early 1970s, making it one of the most consistent longitudinal measures of academic performance available.

The 2025 results, drawn from assessments administered between October 2024 and March 2025 across more than 30,000 students in public and private schools, found that 13-year-olds' scores in both subjects remained below 2020 pre-pandemic levels. There were no statistically significant changes between the 2023 and 2025 administrations for this age group.

The decline for this cohort did not begin with COVID-19. NCES data show that scores for 13-year-olds began to slide in 2012 in both maths and reading, more than seven years before the pandemic arrived.

A separate joint study from research centres at Harvard and Stanford universities, published last month, suggested the drop is part of a broader 'learning recession' that started in 2013 and coincided with a dismantling of test-based accountability systems alongside the rapid rise of social media use among adolescents. Soldner acknowledged at the press conference that 'this isn't just a pandemic story.'

Earlier NAEP data, released in September 2025 by the National Assessment Governing Board, showed that just 22% of 12th graders scored at or above the proficiency benchmark in maths in the spring of 2024, down from 24% in 2019.

That figure of 22% represents the lowest share on record. At the other end of the scale, 45% of 12th graders scored below the NAEP Basic level in maths, also the highest percentage ever recorded. Soldner had warned at the time: 'Scores for our lowest-performing students are at historic lows. My predecessor warned of this trend, and her predecessor warned of this trend as well. And now I am warning you of this trend.'

An International Gap Measured in Grade Levels

The domestic stagnation in maths takes on a sharper edge when set against international competitors. In the 2018 round of the Programme for International Student Assessment, administered to 15-year-olds across 79 countries and education systems, American students scored an average of 478 in maths out of 1,000.

Students in four high-achieving Chinese provinces scored 591 in maths, a gap of 113 points. In 2019, the OECD, which administers PISA, noted that those Chinese students 'outperformed by a large margin their peers from all of the other 78 participating education systems in mathematics and science' and that the most disadvantaged 10% of Chinese test-takers demonstrated better reading skills than the average student in OECD countries.

In the 2022 PISA round, the most recent for which full results have been published, American 15-year-olds scored 465 points in maths, placing the US 34th globally among 81 participating countries and 28th among OECD member nations, below the OECD average of 472.

Notably, the four Chinese provinces that had topped every prior PISA round did not submit comparable data for 2022; the OECD confirmed this was because schools in those provinces were closed during the intended data collection period. Comparable scores from Hong Kong and Macau still placed Chinese-administered jurisdictions well above the US. A full mainland Chinese result is expected to return in the next PISA cycle.

The US average maths score dropped 13 points between 2018 and 2022. Urban Institute analysis of PISA data confirmed that 15-year-olds in the US experienced an 18-point decline in maths between 2018 and 2022, described as a 'substantial' fall. As of the February 2026 National Science Foundation indicators report, assessments continued to reveal 'incomplete recovery for US K-12 students from pandemic-related STEM learning losses.'

The Economic Stakes Behind the Test Score Decline

Federal science officials have drawn a direct line between falling maths scores and long-term economic competitiveness. The NSF's 2026 STEM indicators report found that the US STEM workforce represented approximately 24% of the total domestic workforce in 2023, a sector that drives research output, manufacturing, and innovation. The report identified 'incomplete recovery' from pandemic learning losses as a systemic risk to sustaining that pipeline. In practical terms, what students do not learn in secondary school feeds directly into the STEM degree programmes that produce the scientists, engineers, and technicians the economy depends on.

The Education Recovery Scorecard, a joint 2025 project by researchers at Harvard University, Stanford University, and Dartmouth College, found that as of spring 2024, the average US student remained nearly half a grade level behind pre-pandemic achievement in maths and reading.

The scorecard also documented a deepening inequality within those averages: districts in the highest-income bracket were nearly four times as likely to have recovered to pre-pandemic levels as those in the lowest-income bracket. Professor Sean Reardon, director of Stanford's Educational Opportunity Project, said that 'the slide in average NAEP scores masks a pernicious inequality: scores have declined far more in America's middle- and low-income communities than in its wealthy ones.'

That inequality compounds the international competitive picture. A nationally representative deficit is difficult enough to address. A deficit concentrated in communities that already face the highest barriers to entry into STEM careers means the pipeline problem is not evenly distributed and will not respond to uniform policy solutions.

A generation of American teenagers is leaving secondary school with a maths deficit that no future classroom can fully recover from, and the window for intervention is closing faster than the policy response.